A massive new
catalogue from Maplin
that's bigger and better
than ever before. If you
ever buy electronic
components this is the
one catalogue you must
not be without. Over
300 pages, it's a
comprehensive guide to
electronic components
with thousands of
photographs and
illustrations and page
after page of invaluable data.

We stockjust about
every useful component
you can think of. In fact,
well over 5000 different
lines, many of them hard
to get from anywhere
else. Hundreds and hundreds
of fascinating new lines,
more data, more pictures and
a new layout to help you
find things more quickly.

ifiopum

Maplin Electronic Supplies ltd

All mall to PO Box 3. ►’aylo.jh. Essex SS6 8lR

Telephone Southend |U7D2| 554 1 55 Sales |0702| 55791 1

Shops

159-161 King Street. Hammersmith, London W6 Telephone (0I| 7480926
284 London Road. Westdlff-on Sea. Essex Telephone Southend (07021 554000
Both shops closed Mondays

■ Please send me a copy of your 320 page catalogue.

■ I enclose £ I (Hus 25p p&p). If I am not completely

! satisfied I may return the catalogue to you and have

■ my money refunded. If you live outside the UK send
I £1.68 or 12 International Reply Coupons.

I - UK 03

selektor

canned circuits

bath thermometer (w. Koreio ....

xylophone

can shy (e. Paul)

midnight raid detector id. Butiari .
NiCad piggy bank <w. Hoidinghausan)
a 'flash in the can' ie. Muier) ....

candid comments

drinks round indicator <h.j. Urban)

kaleidoscope IJ.U. Hainqua)

panoramascope (H.Luhmer)

locomotive (H. Thiele)

cumulative index volume 6 1980

dice shaker (r. Mohunioi)

canometer (S. Heilmann)

tilt (P. Groger)

marimba can <w. Frose)

multican IB. de Boer)

mobile can <R. Wenselburger)

dincan (L. van Ginderen)

genie in the can (r. wenuiburgar) . .

piping can (A. Kramer)

simple remote control if. kaiparac)

water can (R. da Boer)

canine defence it.n. stokes)

garden gnome IW. Hoidinghausan) . . .

market

advertisers index

12-01
12-07
12-07
12-09
12-10
12-12
12-13
12-16
12-17
12-19
12 20
12-22
12-24
12-26
12 29
12-30
12-33
12-34
12-35
12-38
12-39
12-40
12-42
12-44
12-46
12-47
12 48
12-49
UK 22

i

The development of traffic
control systems in Glasgow

The City of Glasgow has seen major
developments in traffic control ranging
from relatively simple schemes in the
early 1930s to sophisticated motorway
control schemes in the late 1970s. To
meet the challenge of the 1980s,
Strathclyde Regional Council has
planned a centrally integrated traffic
control (CITRAC) system for Glasgow
which will further enhance the repu-
tation of the city as a pioneer in traffic
control by utilising the most up-to-date
technology.

In 1967 Glasgow became the first city
in the UK to control traffic on an
area basis using a central computer. The
scheme is still in operation in the city
centre and comprises 90 traffic signal
controlled junctions. At the same time
urban motorway construction began in
Glasgow; this took the form of radial
and circumferential motorways and
expressways, designed to carry local
traffic in addition to providing a city
centre by-pass.

The complex network of motorways
is epitomised by the Kingston Bridge,
a dual five lane carriageway carrying
over 110,000 vehicles per day and
containing 11 entry and exit ramps
within a distance of 0.5 mile. To achieve
the highest standards of safety and
efficiency the latest technology has
been utilised to provide Glasgow with
the most sophisticated motorway traffic
control and surveillance systems in the
UK.

SRC has invested £ 1 .5m in these
systems, supplied by two Philips com-
panies - Philips Traffic Systems and
Pye Business Communications. The
Philips association with Glasgow stems
back to the late nineteenth century
when Gerard Philips studied in the
laboratory of the physicist Thomson,
later Lord Kelvin, and supervised the
installation of an electric light system.
The integrated motorway control and
surveillance system is the first micro-
processor-controlled motorway scheme
to become operational in the UK.
It is the latest step towards the CITRAC
system which will eventually link this
sophisticated scheme to an enlarged
surface street area traffic control
system and to a system of diversion
'outes off the motorway.

The sophisticated signalling system,
supplied by Philips Traffic Systems of
Mitcham, Surrey, is a key feature in the
development of Glasgows CITRAC
system. Together with the closed
circuit television system supplied by
Pye Business Communications of

Cambridge, the signalling system will
provide comprehensive control of all
traffic activity on the urban motorway
network.

The motorway control system com-
prises 374 signalling units incorporated
into 97 overhead gantries. The units
can display up to 14 different signals
including flashing red and amber lights
according to legal requirements; each
lane has individual signalling and the
system is capable of controlling 800
signs on 200 gantries.

Two Philips P857 minicomputers, lo-
cated in the computer centre of the
Department of Roads, transmit signals
to the microprocessor controlled out-
stations via party system. The out-
stations are housed in the gantries, and
up to 32 outstations can be attached
to a party line which brings considerable
savings in copper. Each outstation can
control up to eight signal units which
comprise 42 V tungsten halogen lamps
with fibre optic signs; together with an
AC power supply, the lamps will ensure
much longer life from individual bulbs.
The system is operated from the central
control room at police headquarters
in Pitt Street, Glasgow. The operator
can view a schematic representation of
any section of motorway on a visual
display unit, and from its keyboard,
messages are sent to the outstations;
a scrolling facility enables the operator
to scan motorway sections in sequence,
and every action taken is logged by a
teletype for statistical purposes.

A comprehensive set of ground rules,
supplied by the Regional Council's
Department of Roads, has been in-
corporated into the system's software
program owing to the complexity of
the urban motorway network; these
include automatic countdown sequences
which ensure that the motorist is
advised well in advance of restrictions

The system has been designed to facili-
tate ease of maintenance; any fault in
a signal unit is illustrated on the VDU
in inverse video, and each outstations
has a self-diagnostic facility. It is also
possible to test all equipment without
the motorist being aware; engineers,
requiring minimal computer expertise,
can analyse the system without inter-
fering with the signal units and can
program the system easily to accommo-
date changes and expansion. The system
also enables additional data, for
example, information on queues or
weather conditions, to be transmitted
from any outstation to the operator.

The closed circuit television system, also
in operation, comprises 24 Pye 506
CCTV cameras mounted on steel poles
with special wind-down facilities for
ease of maintenance. Each camera is
fitted with a weatherproof housing, a
thermostatically controlled heater/
demister and screen wipers for protec-
tion against the elements.

High quality newvicon tubes have been
specified for the cameras to counter

elektor decamber 1980 - 12-01

the problem of powerful vehicle head-
lamps and street lighting; and each
camera is fitted with 10:1 zoom lens,
variable speed pan/tilt facility and
automatic iris.

The surveillance system is divided, at
present, into three cable networks
covering the Monkland motorway, the
Renfew motorway and the Clydeside
expressways. The system can be easily
extended to accommodate up to six
cable networks, and extra cameras
can be added where required.

At the base of each pole is a control
box with its own built-in micropro-
cessor; power is supplied from a local
mains, and the picture signal is fed via
the central processor to the motorway
control room. The central processor,
located in the computer centre of the
Regional Council's Department of

Roads, is the central element in the

functioning of surveillance system.

Up to four cameras on each of the three
networks can relay pictures simul-

taneously to the control room where
nine Philips 20" video monitors are
controlled by two operators; each

operator has a control panel and a
desk monitor. The panel enables the
operator to select a camera and a
monitor to observe an incident on the
motorway; coloured pushbuttons and a
joystick controller on the panel provide
control of camera functions including
zoom, pan and tilt.

Pictures from all the cameras in one
network can be relayed to a single
monitor in automatic sequence; and a
digital display on each monitor indicates
the location of the camera.

(591 S)

Optical fibre link in service

Another chapter in Britain's telecom-
munications history opened in the West
Midlands recently with optical fibre
equipment being successfully brought
into public service in British Telecom's
network.

The 9 km link (5.1 miles) between
Walsall and Brownhills, capable of
carrying 120 telephone calls simul-
taneously, is the first stage of a new
optical fibre cable network for Britain
which, when completed at the end of
1982, will be the most comprehensive
of its kind in the world. British Telecom,
pleased with progress, is shortly to order
systems for a further network which
will be twice the size of the first.

Five British firms are supplying equip-
ment for the network. BICC and Plessey
have provided the Walsall-Brownhills

12-02 - elektor decer

salektor

system with much of the optical
cable installation work being done by
British Telecom's own engineers. GEC,
in conjunction with Telephone Cables
Ltd, are currently installing a high
capacity link between London and
Reading, capable of carrying 1,920
phone calls of which the first section
(London-Chiswick), has just been com-
pleted. Work by the same companies
on the Oxford-Banbury route is ap-
proaching completion. STC, who
provide both cable and equipment to
British Telecom, is supplying a range
of systems.

In planning this network, British
Telecom expects to achieve several
objectives:

— to speed the introduction of optical
fibres in Britain, as part of its mod-
ernisation programme,

v to ensure that Britain keeps up with
world leaders in this technology,

— to obtain expertise in the design,
installation and operation of a

variety of optical fibre systems, and,

— to give British industry early ex-
perience in supplying optical fibres in

quantity, with consequent benefit for
the industry's export potential.

The new network will eventually use
more than 3,500 km (2,200 miles) of
fibre — enough to stretch from Land's
End to John O'Groats and back, with
some over. It is being made up into
nearly 450 km (280 miles) of cable
which will be installed on 15 routes
and equipped with 34 systems in
England, Wales and Scotland. These
15 routes, ordered last year, involve
more than £6M of investment by
British Telecom.

The follow-on order to extend this
network will be twice the size of the
initial purchase and will require more
than 7000 km of fibre.

An optical fibre is a strand of glass as
thin as a human hair which guides a
beam of laser light, capable of carrying
thousands of telephone calls simul-
taneously. An optical fibre communi-
cation system includes electronic equip-
ment to convert the telephone speech,
into pulses of light, inject these into
the fibre, amplify them as necessary at
intervals along the fibres and convert
them back to the speech wave at the
distant end. Such systems offer signifi-
cant prospects of economic and techni-
cal improvements in running and ex-
tending the nation's telecommunication
services.

The basic raw material, glass, is cheap
and in abundant supply, unlike the
important copper used in conventional
cables. Optical cables are also more
compact than existing types and are
immune to crosstalk and electrical
interference.

British Telecom is part of the Post
Office

Polaroid sonar technology:
ultrasonic solutions to engineering
problems

Picture a maze of corridors in an office
complex, hotel, school building or a
series of narrow aisles in a food market
or department store. Not particularly
adventurous territory for the average
person, but potentially dangerous
settings for the unsighted who must
somehow negotiate obstacles others
can see, from furniture and machinery
to water fountains and potted trees.

Into our imaginary picture comes a
blind student, business executive or
shopper using what appears to be a
white cane - a device traditionally
employed to seek out obstructions
that lie in the path of the user. There
is a difference, however, in that our
subject is using a cane that does not
require reaching out, touching or
making physical contact in any way.
Impossible? Not if the cane is equipped
with a miniature ultrasonic guidance
system, instantaneously and continu-
ously measuring the distance from walls
and other solid objects, then providing
audible signals to establish the location
and distance of potential hazards.
Although such a device is not yet
readily available, it may not be far off.
Polaroid technology now makes it
possible for the unsighted to 'hear'
their way by means of an ultrasonic
echo-ranging system originally devel-
oped and currently used to focus
Polaroid SX-70 autofocus cameras. In
operation, the user activates the battery-
powered system which then emits a
continuous series of inaudible, high
frequency 'chirps' — sound waves that
travel from the electrostatic transducer
or solid objects and back to the trans-
ducer. With the addition of a small
speech synthesiser, distances can be
'announced' by the touchless cane.
Experimental models of the 'touchless'
cane have already been succesfully
tested by various organisations for the
blind.

The remarkable precision of the ultra-
sonic rangefinding system that can make
split-second measurements, even in the
dark, and the simplicity of integrating
the electronics into various systems,
opens the door to a host of exciting,
new applications.

Since Polaroid introduced its SX-70
sonar focusing cameras in 1978, there
has been a flurry of interest in non-
photographic uses for the unique
ultrasonic technology. Spanning a spec-
trum of business, medical, scientific
and transportation needs, the highly
sophisticated system, developed to

instantaneously and accurately focus
camera optics for crisp, brilliant instant
pictures, offers new solutions to old
problems.

How it works

The sonor echo-ranging system is
comprised of several key elements -
an electrostatic transducer, crystal os-
cillator clock, digital circuit and coun-
ter; analogue circuit comprised of a
transformer, amplifier and filters; and
a power or readout section. The trans-
ducer, which acts much like a telephone
receiver, converts electrical energy into
sound and sound back into electrical
energy.

The transducer emits an inaudible
sound - a highfrequency 'chirp' lasting
only 1/1000 second - and radiates the
sound towards the target or in the
general direction being investigated.
Each chirp consists of four ultrasonic
frequencies; 60 kHz, 53 kHz and 50
kHz. The returning signals are accurately
measured and travel time is converted
to distance to focus a camera lens,
turn machinery on or off, provide
audible or metered signals or perform
any number of functions.

Aviation is a promising area for appli-
cation of the echo-ranging system. At
present, there is nothing available to a
pilot that can provide accurate altitude
readings below 20 metres. Commercial
passenger aircraft and small private
planes are finding that exact readings
can aid smooth landings, especially in
poor weather conditions, help flying
low over water, or assist pilot training.
The international press recently hailed
the achievement of Bryan Allen when
he broke the world's record for the
longest human-powered flight by pedal-
ling the Gossamer Albatross across
the English Channel. Not publicised,
however, was the critical need to keep
the craft at a precies distance above
the Channel's choppy waves. This was
accomplised by using a custom-built
altimeter incorporating Polaroid's ultra-
sonic echo ranging system to provide
altitude readings at a rate of five times
per second.

Helicopters, the workhorse of military
and civilian airborne missions, can
benefit in low altitude hover and
touchdown situations. Lowering a heavy
load into a limited access area or other
high risk environment may be extremely
difficult and dangerous for a pilot
concerned with surrounding objects or
unfamiliar terrain. Imagine the ad-
vantages of a continuous series of
precise distance readings while ma-
noeuvring and touching-down on a
moving aircraft carrier, particularly with
critical cargo aboard.

The simplicity of the Polaroid sonar
system, its moderate cost and high
reliability make it possible to equip
large or small aircraft with multiple
transducer's, enabling the pilot to know

precise distances at crucial locations of
the aircraft.

Obviously, there are a great many
aviation-related applications for echo-
ranging systems, safety devices for
baggage loading, precision sensing sys-
tems for mid-air refuelling, controls for
spacecraft docking, and safety mechan-
isms for parachutists to indicate the
last, critical phase of any jump -
knowing how far away the ground is.

Vehicular control and safety
The uses of instantaneous distance
readings are equally limitless for ground
vehicles. The high incidence of accidents
involving trucks and forklifts in factories
and warehouses or on construction
sites illustrates that an automated
rangefinding system can reduce danger
to workers, vehicles and valuable cargo.
The driver who cannot see clearly in
all directions or who is occupied with
mechanical functions as well as con-
trolling the vehicle, can be altered by an
audible signal when the truck, crane,
earth-mover or harvesting machine nears
an obstacle. In fact, the vehicle could
be programmed to stop automatically
when a minimum safety distance is
reached.

Echo-ranging safety can be used for
placing loads in visually inaccesible
areas, backing trailer trucks up to
loading docks, coupling a string of
railroad cars, operating vehicles in
obstructed or congested areas and
safeguarding fragile or dangerous cargo.
For years, automotive engineers, have
speculated on the day when passenger
cars will be equipped with safety
devices to sense the distance from
obstacles in their path — for instance,
when backing into a parking space or
garage. In theory, at least, fewer acci-
dents! Whether or not this goal is fully
achievable, sonar technology has made
the most critical element, the sensors, a
reality.

Echo-ranging has numerous applications
for marine vehicles, too. A typical scene
in any major harbour is the huge ocean
liner or tanker docking in tight quarters,
with a swarm of tug boats frenetically
tooting warning signals. A rangefinding
system could furnish continuous, exact
•eadings of the vessel's distance from
close-range objects - in bright daylight,
heavy fog or total darkness.

Other aspects of shipping can benefit
too. Crane operators loading ship holds
cannot see the unloading site and must
depend on hand signals or radio contact.
Precise distance readouts initiated by a
transducer on the pallet rigging could
provide the information required, in-

stantaneously, and with absolute
precision.

Medical Aids

As helpful an application of sonar
rangefinding as the sonar cane is the
installation of Polaroid's ultrasonic sys-
tem on a wheelchair. Existing tech-
nology makes it possible to automate a
wheelchair to a high degree, enabling
the user to operate the device easily.
The echo-ranging system, with its
continuous transmission and reception
of sound signals, can signal obstacle-
avoidance warnings, provide automated
controls and even stop the wheelchair.

A system that automatically focuses
liquid lenses in a pair of reading glasses
may assist those whose own lenses
have been removed in cataract oper-
ations. The transducer can automatically
measure the distance to the page and
adjust the lenses accordingly.

Precision positioning data can be a
boon to technicians and doctors using
radiation equipment, including X-ray
systems and sophisticated devices like
computer and tomography (CAT)
scanners, when it is essential to maintain
a precise distance between the patient
and radiating elements — for safety as
well as suitable imaging. Ultrasonic
distance measurements can make exact,
continuous calculations that adjust to
body contours, even if the patient
moves. The entire scanning process can
be automated to provide perfect results
every time, lessen the traditional
retakes, unnecessary cost and possible
patient harm.

Applications for ultrasonic measurement
in scientific instrumentation, manufac-
turing and agriculture are literally
limitless. For example, a fruit grower in
Florida has already used the technique
to automate three spraying operations.
The distance readings, from the spray
vehicle to the trees being treated,
control the nozzle aperture, spray
pattern and duration - even turn the
sprayer on and off.

Cybernetics provide a challenging testing
ground for sonar distance measurement
technology. An American teenager won
a science and engineering fair award
recently for his self-controlled mobile
robot. The distance measurements made
possible by Polaroid's echo-ranging
system protected the unit from damage
and allowed it to operate by itself.
Cybernetic equipment, including highly
articulated robots, can provide incalcu-
lable benefits by performing in hostile
environments where humans cannot
function — to handle fires and chemical
disasters, to perform at high efficiency
in explosive atmospheres and in tem-
perature extremes.

In fact, the ability to gauge accurately
and to control remote operations in
potentially unsafe environments is
essential these days. The remote manual

manipulation of nuclear materials, for
instance, could be safer and more |
efficient if operators knew the exact |
distances involved, or if procedures were
fully automated.

For example, think of the remote
transfer of radioactive materials, the
need for exact positioning of fuel rods
in a nuclear reactor, or the handling
of any materials in dangerous surround-
ings, temperature extremes, radioactive
environments, or areas with chemical
fumes and combustibles.

Inventory control is always a major
problem when thousands of items,
many bulky or difficult to count
manually are involved. Want to count a
stack of paper sheets in a high volume
printing operation? Since the number
of sheets per unit of height is known,
the sonar transducer can continuously
measure the height of the stack and
provide information to a calibrated
readout device. Interested in continu-
ously monitoring the volume of liquid
in a huge storage tank — to accuracies
within millimetres of the exact level?
Once again, echo-ranging can provide
the means with sensors located within
the tank.

Whether employed to sense the move-
ment of paper webs on automated
high-speed printing presses, assure elev-
ator stopping accuracy, count objects,
measure and regulate fluid levels or even
guide toys or mechanical robots,
Polaroid's echo-ranging system is certain
to play an important role in bringing
the future measurably closer.

'Talking cars'

Cars of the future are likely to 'talk' to
their drivers.

Microcomputers will make a major
impact on vehicles of the future. How-
ever, they could produce so much
information on the state of the car and
outside conditions that there could be a
problem for the driver in absorbing it
all. This could be solved by the use of
a speech synthesis micro-computer
which would be able to 'talk' to the
driver. The robot passenger could
constantly scan the information pre-
sented by the car's sensors and verbally
inform the driver of any dangerous
situation. The driver could also ask the
talking computer for information such
as average speed and the estimated time
of arrival. A three-chip computer system
already exists to provide a vocabulary of
256 words with adequate diction
quality.

12-04 - elektor decemfaer 1980

The potential of the biochip

Semiconductor technology has brought
microelectronics to the stage where the
size of individual transistors on a chip
approaches that of large molecules in
cellular material. It is reasonable to sup-
pose that the scale will soon be small
enough for us to design circuits capable
of organizing themselves into simple
forms of artificial intelligence. Coupled
with newly-developed techniques for
implanting probes in living material,
they might simulate biological systems
and even be capable, to some degree, of
repairing them.

The idea of implanting electronic
systems into living tissue is not new.
Cardiac devices such as the familiar
pacemaker have now progressed to the
stage where they embody simple micro-
processors to adapt them more closely
to specific characteristics of individual
hearts, and several laboratories are
working on microelectronic hearing
devices that by-pass a defective inner ear
by directly exciting a small part of the
nerve bundle forming the auditory
nerve. These devices use tiny electrodes,
several micrometres in diameter, which
are made by photolithography.

Similarly, it is possible to stimulate a
precise part of the optic nerve, or visual
cortex, to produce bright spots in the
field of vision. By adding a micro-
processor to a multi-electrode system it
might be possible to pre-process and use
them to build up rudimentary images of
the visible world. Other work being
done includes the electrical stimulation
or control of defective neural units in
people who are paralysed in the lower
parts of their bodies or down one side.
Research into the working of the ner-
vous system, including the brain, has
substantially benifited from electronic
techniques for exciting and probing. But
all these developments have been
severely restricted by the lack of large
arrays of ultra-small electrodes and
miniature processing systems capable of
exciting and probing in a fine mesh over
a large enough portion of the neural net-
works, and of doing so without causing
damage. So far, sensors are either too
coarse or too few in number to cope
with the complexity of individual cells
or neural systems.

These problems could, in theory, be
overcome by borrowing from the micro-
fabrication techniques used in making
silicon chips. But there still remains the
problem of transmitting the information
from, say, a 100000-electrode array to
the experimeter. Extensive sorting and
pre-processing is obviously neccessary,
which means incorporating a versatile,
high-density microprocessor. Such an

Figure 1 . Orders of size for features in silicon-chip devices, showing sizes of lithography lines
required.

'intelligent' implantable sensor for
monitoring and controlling might be
termed a biochip.

If advanced medical instrumentation
were to be developed along biochip lines
it could significantly improve our
knowledge of the electrical signals (and
chemical ones, if chemically-sensitive
ultra-small devices were used) that
govern learning, memory and behaviour.
From this knowledge, and by reversing
the job of a sensing biochip to that of
control, it would become a real, if still
distant, possibility that neural tissue
might be at least partly repaired.

Part of the research programme in
advanced microelectronics is to study
the feasibility of such biochip instru-
mentation. The University of Warwick,
in collaboration with certain laboratories
in the USA, are exploring problems to
do with electronic processes in ultra-
small devices, complex system design
and bio-compatible materials important
to future biochip technology. The
silicon-chip revolution is now reaching
a stage where further miniaturization
and circuit cleverness will call for exten-
sive changes in the way devices are
made, in the design of computer archi-
tectures and even in the scientific basis
for understanding and exploiting elec-
tronic processes.

Industry is now getting to the end of
the era of large-scale integrated circuits
(LSI), in which, typically, a micropro-
cessor is manufactured as an array of
some 64000 transistors interconnected
on a silicon chip about four square milli-
metres in area and sizes of the smallest
features are about two to four micro-
metres. Biochip developments will mean
very-large-scale integrated (VLSI) cir-

cuits comprising many millions of com-
ponents packed on to a single chip. I
Individual circuit elements as small as
200 angstroms (20 nm, or 0.02 /jm) will I
be used. This is about the size of large
molecules in cellular matter. Many of
the ideas and techniques of bulk solid- I
state physics which have held for the i
last three decades of electronics will no i
longer work on this ultra-small scale.
Solid-state microcircuits are built by a
process which is basically photographic,
known as photolithography. First, a
wafer of silicon several centimetres in i
diameter is cut from a single crystal.
Next, a thin, insulating layer of silicon I
dioxide is grwon on its surface and i
coated with a photo-sensitive film, I
known as the photoresist. A pattern of
the intended circuit is then projected
onto the photoresist using a beam of I
ultra-violet light or, in advanced litho- I
graphy, some other radiation. The ,
exposed film is then developed by dis- I
solving away the exposed areas, leaving
a pattern of open, insulator regions. .
These are in turn removed by etching,
usually by acids, to reveal the under-
lying silicon surface. Very small |
amounts of impurity atoms may then j
be diffused into the open silicon regions
by placing the wafer in a controlled, hot
furnace containing the impurity gas. In I
this way, the pattern of exposed silicon I
is given the electrical properties that are
wanted. The result is an array of planar
transistors or other circuit elements. A
similar procedure is used to lay down
metallic electrodes and to make inter- I
connecting pathways to link the cir-
cuits.

Photolithography is very economical be-
cause a large number of identical cir- I
cuits can be made on one wafer before 1
it is cut into separate integrated-circuit I

decembar 1980 — 12-05

i i i r

1 1 ;

Doped and undoped silicon regions

Figure 2. Photo-lithographic process for making

the pattern etched; (3) The exposed oxide layer
away and impurity atoms are diffused into the !
an array of planar transistors.

integrated circuits; (1) An oxide layer is grown
e layer; (2) The photoresist is exposed and
is etched away; (4) The photoresist is stripped
lieon layer. The final structure is equivalent to

chips. By making the circuit com-
ponents smaller, more components can
be built into each chip, making it more
versatile. There are additional advan-
tages in that smaller transistors use
lower power and operate faster. Because
the cost per component is related to the
area it occupies on the chip, higher com-
ponent density means greater economy.
The number of components per chip has
about doubled each year since 1960, so
we might expect a million-component
chip to be attained in the early 1980s.

In a move toward VLSI, recent advances
in optical projection systems and the
use of light in the far ultra-violet part of
the spectrum show that circuit features
as small as 0.5 pm can be inserted into
the pattern by photolithography. The
smallest limits are fixed in the end by
diffraction effects, which can be over-
come only by using shorter-wavelength
radiation. Advanced lithography tech-
niques now being developed for VLSI
use soft X-rays and electron and ion
beams to give a resolution of features
down to the order of 100 angstroms
(0.01 pm). New high-precision tech-
niques are being worked on for the
etching and diffusion stages to comp-
lement the fine pattern generation and
transfer; in particular, plasma etching
and ion implantation are promising.
Conventional computer systems will be
difficult to incorporate on VLSI chips
because of the very high proportion of
interconnect paths, which take up a lot
of space. This 'wiring' problem is
brought about because so-called sequen-
tial processing architectures are in use,
in which computations are carried out
as a long chain of logic operations.
Parallel or concurrent architectures, in
which computers can perform a large
number of operations simultaneously.

offer better space-filling path-ways but
are not as well developed. Choosing the
computer architecture to use in VLSI
chips will mean considerable re-thinking
of basic computer science; constraints
imposed by the economies of design and
fabrication of the equivalent electrical
circuits will not be the least of the
problems to be solved.

A great deal of research effort is
devoted to exploring and exploiting
novel electronic processes which be-
come available in sizes somewhere
between solid-state LSI and the true
atomic scale. In a conventional bulk
semiconductor device, electrons or
'holes', which are vacancies in energy
bands normally filled by electrons, are
swept from one electrode to another

through the application of a control
voltage. The time this takes is called the
transit time. A steady average drift
velocity is reached when the rate at
which the electrons gain momentum
and energy from the accelerating elec-
tric field is balanced by the rate of loss
of energy and momentum through
collisions with impurities and vibrating
atoms in the host semiconductor. Ohm's
law is obeyed if the fields are weak and
there are many completed collisions
within the transit time. But in very
small devices the accelerating electric
fields can be very large because the con-
trol voltages cannot be scaled down
below the thermal noise level, that is,
the small voltage arising from the tem-
perature-dependent, random motion of
electrons. Non-ohmic conduction hap-
pens easily and the very short transit
times may make it impossible to achieve
a steady drift velocity. This condition is
called the transient regime.

At smaller scales, the transit time
eventually becomes less than the mean
free time between collisions. The con-
duction becomes ballistic: electrons no
longer see the scattering mechanisms
within the device volume, and free
acceleration should take place. However,
the electrons can still interact with the
encroaching environment of the device,
that is, with the imperfections and
atomic vibrations in the contacts, inter-
faces and surrounding insulator regions.
In this regime, conduction is constrained
by size. In the extreme, when the device
is small enough, Heisenberg's under-
tainty relation indicates that it becomes
more and more difficult to confine the
electrons to the device. Quantum
effects become strong, and the available
energy states for the electrons become
discrete rather than continuous. The
wave-nature of the electron becomes a
dominant factor when the size of the
silicon approaches the 100-angstrom
region, for the electron waves can then

Figure 3. A two-dimensional superlattice. The atomic-like potential barriers and wells ar
electrically controllable by a voltage applied to the aluminium 'gate' electrode. By mean
control voltage V and the superlattice spacing d. the implanted superlattice may be used
over-ride the dynamical effect of the natural silicon lattice.

New system architecture

escape from the device and may overlap
into adjacent devices; this is known as
the tunnelling phenomenon. Conduc-
tion still goes on, but to understand
what is happening requires the full
theory of quantum mechanics.
Superlattices give us a good example of
the sort of quantum effects which can
be exploited in ultra-small systems. In a
perfect crystal, the electronic properties
of the material are fixed by a periodic
array of atomic potentials, which dif-
fract the electron waves as they propa-
gate through the crystal lattice. An
artificial one-dimensional lattice, known
as a superlattice, can be superimposed
on the crystal lattice by growing alter-
nate layers of various materials on a
semiconductor substrate, separated by
a few hundred angstroms. By varying
the composition and separation of the
layers it is possible to control the elec-
tron dynamics in a direction perpen-
dicular to the layers.

Superlatic effects have been demon-
strated by a number of laboratories,
particularly Dr Ray Dingl's group at Bell
Telephone Laboratories in the USA.
The two-dimensional example in fig-
ure 3 is of a superlatic proposed by
Dr R.T. Bate, of Texas Instruments.
This structure could be made by using
one of the advanced lithography tech-
niques.

Co-operative Networks

When the separation between devices
approaches molecular size, it becomes
more and more difficult to isolate any
particular device from its neighbours. In
a similar way to the superlattice
example, the overall architecture of the
VLSI system of devices may become
more important than the host semi-
conductor in fixing the electrical
properties of the array. Already, un-
expected interactions have appeared
between circuit elements. For example,
the phenomenon of 'cross-talk' between
memory cells in high-density LSI
memory chips is accepted as a reliability
problem. Exploiting such behaviour
between devices could lead to more
versatile electronic networks which
would not need the high proportion of
space-consuming wiring patterns that
are now used in microprocessors. The
behaviour of an orthodox electronic
logic system is fixed once the pattern of
devices and their interconnecting path-
ways has been established. A different
behaviour might be imparted to it by
rewiring, that is, re-ordering the devices,
but this is normally impossible in
integrated-circuit systems. We are study-
ing an alternative approach. We have
built theoretical models to simulate
arrays of electronic devices which are
only partially isolated from each other.
The arrays are intended to undergo
spontaneous self-organizing, or co-
operative transitions between differ-
ently-ordered electrical structures. The
idea is illustrated in figure 4, where
groups of devices in the VLSI array are

represented as blocks which communi-
cate through built-in pathways and by
the devices directly interacting.
Information is received at the input in
the form of coded electrical signals
which are processed and passed to the
output as additionally-coded signals. At
the lowest input-signal strengths the
array behaves according to the built-in
architecture. At some higher level of
input signal, cross-interactions between
the devices arise through, for example,
tunnelling of electrons. Competition
between these new channels of com-
munication and the original built-in
coupling then gives rise to a differently-
ordered electrical architecture. The new
architecture, and hence the new pro-
cessing function, is sustained as long as
the input signal stays strong enough.
This type of system has to have a great
deal of freedom in the coupling between
devices, a small number of which
control the others, so many parallel
paths are necessary.

Preliminary studies are encouraging
enough to foresee many applications
for biochips embodying cooperative
VLSI networks. For example, they
might be used in self-healing logic
arrays that would be capable of re-
pairing a certain amount of radiation
damage. They might form the basis of
memory systems capable of sorting and
relating data, and of 'artificial intelli-
gence' units to assist in processes such as
pattern recognition.

Implantable Electronics

Many problems, to do with the materials
used, have to be solved before high-
intelligence, implantable electronic
systems can come about. First, the bio-
chip must be effectively insulated against
saline fluids so that it is not penetrated
by unwanted dopant ions, such as

sodium, which destroy semiconductor
devices. The insulating layer, from a few
hundred angstroms to several micro-
metres thick, must be chemically
bonded to the chip: a conventional
wrapping is incapable of preventing
saline penetration to at least a few
micrometres. Second, the biochip must
be compatible with the host biological
material, so the outer layers must be
made from materials which are chemi-
cally inert, such as plastics. This poses
problems to do with the interface
between the outer, inert layer and the
relatively active electronic layers of the
chip. The layers must stick together well
enough to prevent the structure unpeel-
ing. Third, the phenomenon of electro-
lysis, which might cause the metal
micro-electrodes to dissolve in their
electrolytic surroundings when electric
currents flow, poses long-term corrosion
problems and may make it difficult to
obviate toxic by-products. Though a
number of new materials show promise,
solving the problem of passivating the
biochip implants will not be practicable
for some time.

Bionic science is in its infancy. But in
spite of design and other technological
problems it is believed to be a promising
area for interdisciplinary research, with
many potential benefits to medical
science.

Dr. J.R. Barker,

Spectrum no. 17

(596 S)

1980

12-07

Earlier in the year readers were invited,
in the form of a competition, to send
in their impression of what electronics
could do in a can. No great limitation
was imposed on what type of can was
used although 10 gallon oil drums and
the like were ruled out. This did not
matter however since all entries in-
cluded the use of the good old beer or
soft drinks can.

The ideas poured in and they ranged
from the very funny to the very
serious, useful circuits.

This issue is mainly composed of 23 of
the best entries, in fact, the prize
winners. Judging, which was carried out
by our international editorial staff, was
difficult. Of the final selection, seven
have been published complete with a
printed circuit board. These were
chosen as the best in terms of an original
idea coupled with practicality. These
seven readers will each receive a Casio
LCD credit card calculator that can also
be used to compose and play back
your own tunes.

The remaining sixteen prize winning
entries are again original ideas but were
either not quite so practical or needed
a little extra in the way of construction.
The prize for these readers is the
AM/FM radio Radio shown here.

There is one other prize specially
awarded to the constructor of the
locomotive. Strictly speaking, this idea
did not include electronics, however, it
is fun and we liked it.

What of the other entries? The fact that
they did not win prizes obviously does
not mean that they were no good. On
the contrary, some were hilarious. How
about a Budgie entertainer for example?
So many of the entries fell by the way-
side because the idea (almost always
very good) could not be followed up
with the practical application of the can
itself. For instance, one reader sent in
the complete circuit of an oscilloscope
built in a number of cans. Think of the
problems that they could cause on the
workbench.

We hope you enjoy this issue, it was fun
making it. Who knows, it may supply
that last minute Christmas present. H

bath thermometer

elektor december

The bath thermometer uses sixteen
LEDs to give an indication of three
temperature ranges: yellow ones indicate
'cold', green 'warm' and red 'hot'. The
bath should therefore be at the correct
temperature when one of the green
LEDs is it.

Circuit diagram

The circuit diagram for the bath ther-
mometer is given in figure 1. All the
functions required to control the row of
W. Korell LEDs (D1 . . . D16) are contained in

lialli thermometer

Many people don't like to leave
anything to chance, and often
with good reason. Jumping into a
scalding bath can be a scathing
experience, to say the least! The
more tentative amongst us will
therefore welcome this bath
thermometer which can be used to
test the temperature of the water
before taking the plunge. It will
satisfy even the most demanding
of bathers from infant to
grandmother.

IC1. The 'scale' of the thermometer is
determined by the voltage levels on
pin 12 and pin 13 of I Cl. The upper
threshold (pin 13) is fixed at a level of
approximately 5.2 V while the lower
threshold (pin 12) is adjustable by
means of the preset potentiometer PI.
If the input voltage at pin 11 is outside
the preset limits, either the first or last
LED in the row will be on continuously,
depending on whether the input voltage
was too low or too high respectively.

The input voltage is determined by the
resistor network P2, R3 and the NTC
(temperature dependant resistor,
NTC = negative temperature coef-
ficient). The input voltage will therefore
vary as the resistance of the NTC varies
with temperature. The exact tempera-
ture can be established with preset
potentiometer P2. Two small 9 V
batteries are all that are required to take
care of the circuit's power supply. The
voltage at pin 10 of IC1 is stabilised
(12 V) by IC2. Resistor R1 determines
the brightness of the LEDs.

included for another application of the printed circuit board.

2-08 - alektor decembar 1980

Figure 2. The side view of the inside of the
bath thermometer shows the principle
components involved in the construction.

Construction

How do we get it all into the can? First
the components are mounted on the
printed circuit board shown in figure 3
and the circuit tested. The Candid
comments article gives some advice
on how to open the can. The top
should be removed in such a way that
the folded edge remains undamaged, if
possible, as this will be used as part of
the watertight seal. One or two holes are
then drilled in the base of the can
(depending on the type of thermistor
used) for the connecting wires of the
NTC. The NTC can then be mounted
using a suitable epoxy resin in such a
way that the holes and the thermistor
are covered (test for leaks with waterl).
When the can is completely waterproof,
a few ballast weights are stuck to the
bottom from the inside (see figure 2).
The printed circuit board, the perspex
lid and the battery holder (if required)
can be combined to form a slide-in unit.
The batteries could even be attached to

R1,R4 = 10 k
R2 = 1 k
R3 = 1 k5

PI = 10 k preset potentiometer
P2 = 5 k preset potentiometer
NTC = 20 k thermistor

Capacitors:

Cl = IOp/16 V

Semiconductors:

D1 . . . D5 - yellow LED
D6 . . . D1 1 = green LED
D12 . . . D16= red LED
IC1 = UAA 1 70
IC2 = 78L12

Miscellaneous:

SI = miniature spst toggle switch

the base with double-sided tape.

In either case, however, the circuit
board and the board (or perspex disc)
on which the LEDs are mounted will
have to be spaced from each other by a
distance corresponding to the length of
the LEDs. Apart from this, the perspex
lid has four holes drilled into it for the
screws of the clamp attachment and one
in the centre for the screw which holds
the complete assembly. For the clamp
attachment it is best to use screws with
retaining clips with hold the slide-in unit
firmly in place beneath the lid edge. It is
sealed with a rubber band or O ring in
the groove. Finally, a miniature toggle
switch is mounted in the perspex lid.
This will also have to be made water-
proof with the aid of a rubber band or
0 ring.

Once the thermistor, switch and batteries
have been connected to the circuit
board, the bath thermometer must be
tested again and any adjustments
required can be carried out with the aid
of PI and P2. Nominal bath tempera-
ture is within the range 24°C . . . 38°C.
Now everything can be put together and
the thermometer taken for its first 'dip'.
If the can is too buoyant, or floats at a
peculiar angle, extra ballast weights will
have to be added or the alignment of
the slide-in unit and/or batteries
re-adjusted.

As can be seen from figure 3, the
printed circuit board for the bath
thermometer is of a somewhat multi-
purpose design. The inner circle of the
component overlay has the same
diameter as the can. The outer circle,
however, corresponds to the diameter of
the majority of rev counters on the
market. Of course, a rev counter could
also be built into the can. A suitable
control circuit was published in Elektor
way back in 1976, but it is too large for
this purpose. If a rev counter is required,
it should be possible to construct the
circuit on a small veroboard. H

xylophone

With Christmas practically on the doorstep, small, sound producing
circuits become very popular. This novel circuit, although not entirely
original, has been included mainly because of its simplicity and, by
making use of the printed circuit board, can be constructed very
quickly — just in time to replace that forgotten Christmas present!

This design again makes use of reed
switches, only on this occasion they
have a different function. By placing
several of them next to each other, a
form of keyboard is obtained which
can be played by means of a magnet
attached to a wooden stick.

The circuit diagram of the xylophone
is shown in figure 1 . Eight reed
switches are used to make up a single
octave. Closing any one of them will
start up the oscillator. This consists
of N1, Cl and the resistor in series
with the closed reed switch. As the value
of Cl is fixed, the actual frequency of
the oscillator is determined by the
combined value of the particular resistor
and series potentiometer.

The frequency of the first note, C, is
fixed by the values of Cl and R1 and
will be about 350 Hz. Potentiometers
PI . . . P7 are used to preset the fre-
quency of the other notes to the accu-
racy required. The output signal from
the oscillator is buffered by N2 and
amplified by the darlington transistor
T1 to provide sufficient drive for the
loudspeaker. The volume of the output
signal can be adjusted by means of
preset potentiometer P8. The reed
switches should be positioned approxi-
mately 1 ,5 ... 2 cm from each other on
the board.

To play the xylophone a small magnet
is mounted on the end of a wooden or
plastic rod. When this is brought suf-
ficiently near to a reed switch the latter
will close and start the oscillator.

There is of course one thing to bear in
mind when playing the xylophone:
don't hit the keys as hard as you would
if you were playing a 'real' version of
the instrument, as the reed switches are
encapsulated in glass! H

12-10 — elektor december 1980

can

A well known favourite stand at any
fair used to be the coconut shy where
a crowd invariably gathered to watch
the occasional expert knock down too
many coconuts to carry. Regrettably,
this atmosphere is becoming a thing of
the past with slot machine and pin table
booths being a poor substitute. It's still
possible, however, to try your hand at
knocking down skittles or other objects
with a set of five, well worn, wooden
balls. The design in this article is based
on just this subject. Can shy is a game
consisting of a few cans, a few soft balls,
some electronics and a little effort.

that the can is 'knocked down' and the
player can concentrate on the remaining
five.

More fun is added to the game if each
can is made to look like someone you
know, such as Dad, Mum, M.I.L., your
local traffic warden or even the prime
minister. You can then have a knock at
them without offending their feelings, I
or better still, without them hitting I
back.

A word or two on construction

Dealing with the construction first and
the circuit later may seem a little

E. Paul

Why the electronics in such a simple
game you may ask. After all, the knock-
about cans in the oldfashioned shies
managed very well without. That is true
but they had to be replaced by hand.
Include a few components and the tire-
some task of gathering up the fallen
cans and returning them to the shelf is
removed. In addition, electronics helps
to make the game a good deal more
interesting. Read on to find out how.
Each can is fitted with two eyes (two
green LEDs) and a nose (two red LEDs).
The eyes light up individually for a
short period, one after the other in each
can. The player can now only 'knock
down' the can with illuminated eyes. If
the target is hit, the eyes will go out and
the cans 'nose' will turn red. This means

unusual, but in this instance it is more
practical as the circuit is based on
various mechanical aspects.

With the aid of rubber bands and a few
'screw eyes', the cans are suspended in a
chipboard (or similar) frame in such a
way that they are able to swing freely.
The illustrations in figures 1 and 2 give
a good indication of what it should look
like. It is important to stabilise each can
to prevent it from twisting on its own
axis. Using three mounting points on
the top and bottom of each can would
accomplish this.

A set of six contact plates must be
fitted on the rear wall of the frame and
positioned to line up, as near as poss-
ible, with the centre of each can. Suit-
able material for the plate would be

Figure 1. Front view of the can shy.

copperclad board. This must be c

pin (a screw for example) is fitted ii

such a way that a gap of not more than
3 cms exists between the end of the pin
and the contact plate. It should be
noted that the 'pin' must be insulated
from the body of the can.

When a player hits a can at the right
time, the can will be knocked back to
make a contact between the pin and
the plate. Optical evidence of the bull's
eye will be seen by the nose lighting
up and the eyes going out. To prevent
having to replace cans after every fourth
or fifth throw, it will be advisable to
use a very soft ball and throw gently.

The circuit diagram
As can be seen from figure 3 the circuit
diagram is divided into two sections. To
the left (section A) is the hit recognition
circuit and to the right (section B) is the
clock generator and can selection cir-

First let us consider section A, the hit
recognition circuit. Six of these must be
constructed and one placed in each can.
The circuit basically consists of two
AND gates built up with the NANDs
N1 . . . N4. Provided pins 2 (N2) and
10 (N4) are high and that the can con-
tact (SI) is open, the output of the first
AND gate at pin 6 will be at zero. From
this it follows that pin 9 will be low and
the output of the second gate (pin 11)
will be high. The transistor T1 therefore
switching on the eye LEDs, D3 and D4.
The circuit will then be ready to indi-

be high together with the corresponding

output at pin 6. A trigger current will
then flow via resistor R1 to the thyristor
gate which switches on and activates the
nose LEDs. This causes the input of the
second AND gate pin 9 to go low which
is in turn transferred to the output
pin 11 causing transistor T1 to put out
the eye LEDs. The hit display is acti-
vated and will remain lit until switch S2
is depressed to interrupt the supply volt-
age. The circuit is now ready and
waiting for the next hit.

If by now all the mechanical and elec-
trical tasks have been carried out, the
first ball can be thrown. The six cans
can be 'knocked down' one after the
other, switch S2 can be depressed and
the entire game can start all over again.
Section B is what makes the game all
the more interesting, for it makes sure
the eye LEDs only light for a certain
time. The clock generator is built with
the 555 timer 1C operating as an astable
multivibrator. The duty cycle can be
varied with the aid of trim poten-
tiometers R9 and R11. The pulse
duration - the time during which the
eye LEDs are lit - is set with the
trimmer R9 and the pulse interval with
R2. The output signal of the clock
generator (pin 3 of the 555) feeds the
4017 decade counter. Of the ten out-
puts only six are required for the six
recognition circuits. Each output is con-
nected to one input of the two AND
gates in the recognition circuit (pins 2
and 9). Now the target will move from
can to can at the clock frequency.
Which counter output is connected to
which recognition circuit is left to the
cans owner to decide. The order can be
changed from time to time, which pre-
vents experts from walking off with all
the prizes. N

3

12-12 - elektor decamber 1980

midnight raid detector

D. Butler

Although the age of Viking invasions
has long since passed, a certain amount
of small scale plundering still goes on. In
particular, the nocturnal wanderings of
certain members of the family on their
way to raid the fridge, comes to mind.
Upon investigation the following
morning a few crumbs scattered on the
plate is all that remains of the large
chunk of cheese you were saving, the
last drop of your favourite beer has
mysteriously evaporated and there is no
sign of the culprit. One deterrant would
be to lock the fridge, another to elec-

and capacitor C3, form a squarewave
oscillator with a frequency of around
1 kHz. This oscillator is turned on and
off by a second comprised of N1,N2,
R2 and C2. The frequency of this
second oscillator is much lower than the
first, approximately 3 Hz. It can be
started and stopped by means of the
pushbutton switch SI. When SI is
closed. Cl will charge up almost to
supply voltage. When SI is open. Cl will
slowly discharge through R1 ensuring
that the circuit remains operative for a
few seconds longer.

midnig ht raid detector

This canned alarm is intended to
repel inmates who insist on
delving into the fridge late at night
in order to quench their thirst or
make off with a tasty snack. It can
also be used as a deterrant for
'extra-curricular' snacks by
persons on a diet.

trify it, but these strike us as being
rather extreme forms of punishment.
There are kinder and more humane
ways in which to cure the compulsive
raider and the circuit described here
provides a useful example.

Picture the scene. It is the dead of night,
pitch black and all around is as silent as
the grave. Suddenly, a ghost in pyjamas
creeps stealthily downstairs. Who is it?
Maybe dad carefully groping his way to
the kitchen. The fridge door creaks
open, the light flashes on and a hand
moves in the direction of the beer cans.
The nearest can is grabbed and . . . bleep
bleep bleep bleep ... a horrendous din
shatters the silence. There is no need to
relate what happens next!

The circuit diagram of the contents of
the can is shown in figure 1. A single 1C
and two transistors are the only active
components required. The NAND gates
N3 and N4, together with resistor R3

The output of the 1 kHz oscillator is fed
to transistors T1 and T2 which are
connected as a darlington pair and
which provide sufficient signal amplifi-
cation to drive the loudspeaker. Capaci-
tor C4 serves to decouple the entire
circuit. As CMOS gates are used, current
consumption will be very low, even
when SI is closed, and a small 9 V
battery is the only power source
required.

The choice of can into which the circuit
is to be incorporated will obviously
depend on the particular brew that is
normally consumed. A hole for the
loudspeaker is made in the base of the
can as shown in figure 2. The loud-
speaker used must naturally be smaller
in diameter than the can itself. Enough
space must be left to mount the push-
button which will protrude slightly
further than the base of the can. The
circuit board for the unit can be made

1

NiCad piggy bank

amber 1980 - 12-13

J?

iiiml

t I

rfggy hank

W. Holdinghausen

The relative shapes of a can and nickel cadmium cells could lead to the
porcine conclusion that they were related. We are not sure how we got
into this, but the friendly looking pig is not only an ideal container for
the baby NiCads, but it feeds and maintains them in a healthy
condition.

Figure 2. This drawing shows where to
position the loudspeaker and the pushbutton
on the base of the can.

small enough to pass through the
loudspeaker hole. To prevent the
contents of the can rattling the can can
be padded with foam rubber.

The battery and pushbutton should be
connected to the circuit board and then
the whole assembly can be inserted into
the can via the loudspeaker aperture.
Once the pushbutton has been mounted
in the hole provided, the loudspeaker
can be installed by glueing it to the base
of the can. Battery replacement will
then become rather tedious, but it
would be more difficult to construct
a removable lid. In any case, one battery
should last for a considerable length of
time.

if all goes well the circuit should
produce a noise as soon as the battery
sn<j pushbutton are connected. When
tie can is put down, with the loud-
speaker on the bottom, the noise should
r.op after a few seconds. If it doesn't,
pushbutton may not protrude far
though and will have to be adjusted.

= _ al note: If no-one in your family
-aopens to be a phantom raider, the
c rcuit can still be employed at parties
.-.-ere a much larger audience will be
oresent! M

How did the subject of pigs get into
this you might well ask. A fair question
to which we are still looking for a
reasonable answer. However, we did ask
for canned circuits with particular
emphasis on the unusual . . .

Mother pig is, in fact, a rather sophis-
ticated NiCad charger. When first
switched on, she checks to see how
hungry her NiCads are. If necessary they
are supplied with a charging current for
a period of time. The current is then
reduced but remains at a high enough
level to ensure that the cells are main-
tained in good condition.

Operation

Mum contains quite a few vital organs.
In the first place, the supply is stabilised
by means of a voltage regulator IC5.
Next, the circuit round IC1 monitors
the voltage fed to the NiCads. Charging
time is taken care of by IC1 and IC2. Of
course, being a mum, she also has to
have a memory. This consists of N1 and
N2 and ensures that the NiCads can
only be discharged once.

Keeping a close check on the supply is
carried out as follows: when the supply
voltage is switched on and the batteries
are connected, IC1 monitors the voltage
of the cells by way of R15. If this volt-
age is high enough (the cells are there-
fore not quite empty), pin 7 will become
low. This causes the voltage across C5
to rise very quickly, whereas the voltage
across C6 will rise at a much slower rate.
As a result, transistor T3 will discharge
the NiCads via R16.

As soon as the cells are empty enough,
(depending on the setting of PI) pin 7
will become high and C5 will discharge
within a matter of seconds via R7. This
causes flipflop N1/N2 to change state
and stops the discharge. Pin 3 of IC1 has
the same waveform as pin 7 and so will

Details concerning the mother

* It is suitable for four cells.

* It discharges to a set voltage.

* It charges to a set voltage or
for a set time period.

* Its maximum charge current is
100 mA.

When dealing with this particular elec-
tronic pig, it is as well to remember that
when the green eye is lit she is relatively
happy with her brood. But when her red
eye lights — she will charge!

also be high now. This causes a small
current to be fed to the NiCads via the
constant current source, T1, which
charges them. While they were being
discharged the green LED was lit, now
the red one will be lit.

When the supply voltage is switched on,
the time delay circuit around IC2 and
IC3 will have been reset to zero via C8.
After about 14 . . . 17 hours (adjustable
with P2) output Q7 will go 'high' and
this will reach the input of IC1 via D9.
This changes the state of outputs 3 and
7 to zero again. The red LED will go out
and charging will stop. The green LED
will light, but the flipflop N1/N2 will
now be unable to change, because the
voltage across C6 will no longer be zero.
The batteries will therefore not be dis-
charged either, and because they always
discharge a little by themselves, anyway,
they will be supplied with a small
current by way of R 1 to compensate for
this.

Thus, to put it briefly, the green LED
lights and the batteries are discharged.
Then the green light goes out and dis-
charge stops. The red LED will now
light and the batteries are charged again.
When they are fully charged, the green
LED will light again.

R8.R19- 1 M
RIO- 27 n
R 1 1 = 560S7
R12- 47 Jl/'/iW
R1 7 = 220 k
R18 = 680 k
PI - 10 k preset pot.

P2 = 250 k preset pot.

Capacitors:

Cl -470 p/35 V
C2- 47 p/16 V tantalum
C3.C7.C8- lOOn
C4.C6- 10p/16 V
C5 - 100 p/16 V

Semiconductors:

T1 ,T2 = BC 557B

T3- BC 54 7B or BC140

D1.D2- 1N4001

D3= LED. green

D4- LED, red

D5. . . D9- DUS (1N4148)

IC1 = 555

IC2 - 4060

IC3 = 4040

IC4-4011 B

IC5- 78L12

Miscellaneous:

transformer 2 x 1 5 V 100 mA

Construction

Building the printed circuit should not
present too many problems, leaving
all the more time to 'decorate' the can.
Figure 3 shows what is meant. It is
not compulsory to copy this idea, of
course, for it is much more fun to give
free reign to your imagination. Just
make sure, it's electrically safe - there
is no need for the pig to be suicidal -
insulate the mains transformer and
ensure a good earth connection. The top
of the can is left untouched, this is
where the two LEDs and T3 are
mounted. For the sake of appearance it
is more amusing to use a BC 140 for T3
rather than a BC 547. Their connections
are identical. Test the electronics before
assembling everything in the can.

Setting up

Two different methods are possible. The
simplest method makes sure that the
batteries are not discharged too much,
lower than one volt for instance. Poten-
tiometer PI is turned until the circuit
stops discharging at 4 volts. Do this by
turning PI clockwise. This will mean

that charging will stop at a point several
volts higher than the switch off voltage.
However, this is so high (6 to 7 volts)
that the NiCads will never make it. This
is why charging is stopped after a set

The other method of adjustment does
not set the discharge voltage, but the
voltage at which charging is stopped.
This should be at 5.7 volts (1.42 V per
cell) and is set by PI at room tempera-
ture. The discharge voltage will then be
about 3 volts. If the set switch-off volt-
age is not reached within 14 hours,
charging will be stopped after this
period anyway by the timer.

Something should also be said about the
maximum charge time. This time is
counted from the moment at which
mother is connected to the mains. If the
NiCads are first discharged, the remain-
ing charge time will be shorter than the
set time. It may therefore be better to
extend this time a little.

It is also possible to shorten the dis-
charge time by selecting a lower value
for R12. Then, however, the maximum
charge current will be less than 1 00 mA.
The circuit in figure 4 is used to replace
the four NiCads during the setting up.
It consists of a normal stabilised supply
with an adjustable output. The resistor
connected in parallel across the output
ensures that the voltage does not rise
while setting the charge rate. It is in-
cluded because most stabilised supplies
can only produce current and not
consume any.

To assist with the setting up of the
14 hour period, it is useful to know
that the input of IC3 (pin 10) should go
high in just under 7 minutes. H

-

12*16 — i

iber 1980

flash ir

E. Muller

The only electronic components re-
quired for the construction of the
'rifle' are two nine volt batteries, an
electrolytic capacitor, a low voltage
light bulb and a lens with a focal length
in the region of 30 ... 50 mm.

The target consists of a simple optical
schmitt trigger with an adjustable
threshold so that it can be made to
operate correctly despite variations in
ambient lighting conditions.

The circuit of the "light emitting pistol'
is shown in figure 1. Two 9 V batteries
connected in series charge a capacitor.
When the microswitch (S) is closed the

a flash in

die can 9

Most of us will never admit to
being a poor marksman when
confronted with the rifle range at
the local fairground. Of course,
we are all experts — or at least we
would be if it was possible to
practice at home (Thankfully, the
legal eagles of this country take a
very dim view of people popping
off their relatives and neighbours! ) .
The circuit described here,
however, is perfectly harmless and
allows you to construct your very
own shooting gallery.

capacitor is discharged rapidly through
the low internal resistance of the bulb.
To enhance the intensity of the 'flash',
three times the normal operating voltage
of the bulb is applied for a split second.

1

Figure 1. The circuit of the light emitting
pistol could hardly be simpler. The fully
charged capacitor is discharged rapidly
through a low voltage bulb.

In addition, the light is concentrated
into a beam by a small bi-convex lens. It
is therefore important to mount the
bulb at the focal point Of the lens.

An example of possible pistol construc-
tion is shown in figure 2. The micro-
switch is mounted so that it will be
operated when the pistol trigger is
pressed. A closer approximation of the
true rifle range may be obtained by
providing the pistol with an external
power source instead of the 9 V battery.
Each marksman can then 'purchase'
a certain number of rounds (capacitor
charges) from the range warden (which
of course will provide you with extra
funds to replenish the gallery should
any cans become damaged — wishful
drinking?).

Figure 3 shows the circuitry involved
in the target unit. It consists of an
optical schmitt trigger which is con-
structed from two transistors and an
LDR (light dependent resistor). Initially,
the circuit is adjusted so that normal
daylight will have no effect on the
schmitt trigger. That is to say, the LDR
(R1) will have a relatively high resist-
ance and transistor T1 will conduct.
As a result, transistor T2 will be turned
off and the two LEDs (D 1 and D2)
will be unlit. If a more intense beam
of light (from the pistol) strikes the
LDR, the positions will be reversed. The
voltage drop at the base of T1, caused
by the fall in resistance of the LDR,
will turn off this transistor, T2 will
turn on causing the two LEDs to light.
This situation will remain until the
resistance of the LDR increases once
more.

The complete circuit of the target can
be inserted into an empty beer can
without to much difficulty (it's just a
question of finding somewhere for the
previous contents . . .). The only im-
portant issue is that the LEDs and the
LDR must be mounted so that they
face the marksman (see figure 4).

a flash in the can

T1.T2 ■ TUN IBC 107. BC 547)

Figure 3. The circuit diagram of the target
consists of an optical schmitt trigger. When an
intense beam of light falls on the LDR the
LEDs will light.

Figure 4. The LEDs and the LDR should be
positioned so that they face the 'marksman'.

The preset potentiometers PI and P2
are adjusted so that the ambient lighting
just fails to operate the schmitt trigger.
Potentiometer PI serves as a 'coarse'
control and P2 as the 'fine' adjustment.
Of course, if the surroundings are too
well lit then it may be impossible to
adjust the unit so that the LEDs go out.
The solution to this is to turn a few
lights off (or if you are outside and the
sun is that bright, what are you doing
playing cowboys??).

As soon as all the manual adjustments
have been completed, the game can
start. The distance between the target
and the shooter should not be more
than about 10 m. When the pistol
$ aimed at the target the light beam
should strike the LDR to register a
"hit', indicated by the LEDs lighting.
To reactivate the target the LDR will
have to be covered (shaded) for a short
while. Just enough time to empty the
next target container! H

elektor

1980 - 12-17

Fortunately, beer cans do not come in
all shapes and sizes. The average size is
almost the only size. All, without
exception are round, most are about 5
or 6 inches high and seemingly, almost
all eventually find themselves in the
local park or picnic area.

Before delving into the real depths of
can opening, it would be as well to first

candid
coin infills...

. . . on opening cans

How to open a can.

Yes, we know that you grab the
ring in the top and heave, but a
problem arises when you try and
push a 2 x 3 inch printed circuit
board and assorted components
down through the hole it leaves.
Not to mention transformers,
sheets of perspex, hanging
weights, motors, loudspeakers,
relays and a whole host of LEDs.
As you will find, in the pages of
this issue, all these things and
more have to, somehow, find their
way into the average sized can.
Read on to find out how simple
it can be.

make a pointed comment on what not
to do. We at Elektor have this habit of
trying to keep as many of our readers
as possible. So if you are reading this
with an Aerosol or other spray can in
your hand with a view to cutting it up,
please - please put it in the nearest
rubbish bin. These cans are not for
opening, even in the fourth category —
impossible (more on this later). For the
record, many of these cans, even when
empty, still retain a fair amount of
pressure. A warning is printed on them,
please take note of it.

On now to less dramatic cans. First get
your can and empty it. Beware, this is
the part that you could get to like.
Remember that cans are available in a
great variety of colours and the wrong
one could spoil the finished effect you
have in mind for your project.

There are obviously a number of ways
to open a can besides the one already
mentioned. These methods fall into
three categories. Difficult. Messy. Very
difficult and messy. There is also one
other but that is impossible. Joking
aside, all the methods are admirably
suited to cutting fingers so do be very
careful.

It is equally difficult to judge which is
the easiest or which is the most difficult
method. It's all probably subjective
anyway. We will start with the difficult
method first.

The difficult method

This involves cutting the bottom or the
top out of the can. To do this, a ring of
holes must be drilled in whichever end
you wish to remove. First, mark a circle
slightly in from the edge of the top
(or the bottom) of the can. Next, with a
centre punch (not a screw driver) make
a series of 'pop marks' around the circle,
spaced apart by just over the diameter
of the drill that you intend to use. The
drill should not be too big, by the way,
harm can come to a young lad like that.
Then, very carefully, drill out the ring
of holes. Don't worry too much about
the appearance of the result, unless its
covered with a red liquid, that could be
blood! All that remains to be done now
is to, very carefully and slo»'ly, file the

12-18 - elektor de

ar 1980

candid comments

ragged edge smooth. Patience here will
give an excellent result.

The messy way

This method can be a lot of fun. How to
open a full, unopened, can. Well for a
start you need will power because the
contents will get wasted. Soft drink cans
will be fine for this method. Lesson one,
do not SHAKE the can, never mind
what we tell you later. Lesson two, if
you own a sou'wester and wellies, put
them on.

Obviously, you use this method when
you want to keep the top of the can in
its original state. In other words, your
project simulates an unopened can, the
Midnight Raider Detector for example.
First, take yourself, the can, a small nail
and a hammer outside. Now, place the
can upside down on the ground, hold
the nail with its pointed end touching
the centre of the bottom of the can and
- wait for it — hit the nail with the
hammer. For added family entertain-
ment, first shake the can vigorously for
five minutes. Hilarious isn't it. Once the
hole has been enlarged, the contents can
easily be poured out. Carry on using the
Difficult method remembering to epoxy
the opening tab from the inside.

The very difficult and very messy
method

Now we come to the hard part, cutting
the can in half. Seasoned men have been
known to go glassy eyed at the very

thought of it but, provided you have
emptied enough cans to practise on, you
should be all right. Count them, if you
have over ten forget it and try again
tomorrow. For those hardy souls still
standing we will continue.

The trick in this one is to stop the can
from buckling in the process. First, fill
the can with fairly damp fine sand (or
better, cement powder, but don’t take
all afternoon about it) and tamp it
down by banging the can down smartly
onto a solid surface. Try the ground,
covered with a thick piece of card to
save damaging the bottom of the can.
Keep topping up the can until you can
feel the sand is solid through the hole
in the top. Don't forget to keep the
sand nice and damp. Fun isn't it. When
you are satisfied that no more sand will
go in, mask the hole with masking tape
or similar.

A layer or two of the same masking tape
can now be wound round the can at
approximately the height at which you
intend to cut. This serves two purposes.
It will (hopefully) protect the finish on
the can should the saw slip and it will
allow you to mark a guideline easier.
The latter can be carried out by stand-
ing the can on a flat surface. Then take
a pencil and hold it against a solid heavy
object placed next to the can. Put the
point of the pencil against the can (at
the right height), and rotate the can.
Simple? Try it.

Having masked, marked and packed the

can, it can now be cut using a very fine
saw blade, a Junior hacksaw would be
suitable. What do you mean — the sand
all runs out? Back to the beginning
again. Go and get another can from the
fridge (and empty it) and this time
dampen the sand more and get more in.
One little variation on the above theme.
We didn't want to tell you, but try using
plaster of paris, instead of sand, and
allow it to set before cutting. This will
result in total success but it isn't half
the fun! How do you stop the plaster
from sticking to the can? Easy, swirl it
round with a drop of oil first.

Finishing

There are probably a great many
methods of covering the ragged edges of
the cut can to give it a neat appearance.

A simple and cheap way out is by using
a length of PVC insulating sleeving, slit .
down one side, and glued over the edge.
It's useful to remember that it is avail-
able in a variety of colours. Nice.

Both the Kaleidoscope and the Genie
require the use of two sections of a can
fitted together. To make the joint well
needs quite a lot of mechanical skill.
The use of a 'sleeve' between the two
halves is probably the easiest — if you
can find a suitably sized can to make
the sleeve from.

To make the can exactly the same as
when you bought it, watertight that is,
also needs some ingenuity. It really
depends on what you can find in your
junk box in the form of a sealing ring.

If any readers do happen to know of a
sealing ring that will fit the top (or
bottom) of a can, please let us know,
give us a ring so to speak.

An important point to remember. Many
cans are sprayed internally with a type
of plastic during manufacture. This film
makes a good insulator but should not
be relied upon where mains is con-
cerned. Make absolutely certain that the
mains earth point does actually reach
the metal of the can and make any
insulation in the proper manner.

Before fitting your project into the can j
it might be as well to remove all traces
of the previous contents, be it beer, soft
drink, damp cement powder or sand.
Not one of these materials can in any a
way improve on the electronics and
besides, traces of three month old beer
emanating from your Bath Thermometer
might cause questions to be asked. 1
We don't pretend to have covered all
aspects of can opening but we have tried ’
to give a few less obvious pointers.
There are not too many reference works j
on the subject and up until this issue the
editorial staff knew of only one method
of opening a can. In fact, they practise
it regularly.

A prize will not be awarded to the first
reader to count how many times the
words 'can' or 'cans' is mentioned in j
this article. Please mark your completed |
entry, 'cans' and send it to Terry Wogan,
c/o BBC, London. M

drinks round

elektor de

sr 1980- 12-19

H.-J. Urban

Figure 1 gives a general idea of what the
rounds indicator looks like. Nine LEDs
are evenly spaced around the can's
circumference, one for each drinker.
The indicator operates rather like an
electronic version of Russian roulette,
except that instead of risking a bullet
through your head you risk being
carried home with an empty pocket,
but full of beer.

Once the start button has been de-
pressed all the LEDs will light up
faintly. When the pushbutton is re-
leased, all but one will go out. The first
victim will be indicated by the remain-
ing LED shining brightly in the dimness
of the bar. Which LED this is depends
entirely on chance as a form of random

generator has been incorporated.

The circuit operates as follows. First
the on/off switch, S2, must be closed.
The oscillator constructed around gates
N1 . . . N3 will then generate a series
of pulses which are fed to the clock
input of the 4017 counter, IC2. When
switch SI is closed, these pulses will also
be directed to the re-triggerable mono-
stable multivibrator formed by IC1.
The 0 output of the monostable will
then go low as will the clock enable
input (CE) of the counter (IC2). The
counter then starts to count. When the
pushbutton is released the 0 output of
the monostable will go high again after
about 2 ... 3 seconds and the counter
will be stopped. The state of the count

d rink s round
indicator

Games of skill and games of chance have always been popular pastimes
in the social surroundings of the village local, and always will be
(providing the brewers slow down on their price increases!).

These games are often played to determine who is to be the next
(unlucky) person to have to pay for the round. The playing of these
games (darts, dice, dominoes etc.) is not only energetic but they also
tend to take up a fair amount of time that could be put to better use . . .
The drinks round indicator described here provides an interesting
alternative to traditional methods since it keeps the beer flowing by
electronic means (how you get home afterwards is of course a technical
problem which is beyond the scope of this article).

Figure 1. A suggested ‘layout' for the can.

2

kaleido-
scope

C1.C3- 10 (i/16 V
C2 - 3n3

We never fail to be amazed at the
things that some of our readers
can produce from virtually
nothing in terms of bought
components. A good idea for
the basic project is obviously
a head start. The kaleidoscope
in this article is an excellent
example of something practical
and original from next to nothing.

D10...D18-LED
IC1 =4528 or 4098
IC2-4017
IC3-4011

at that moment will be indicated by
the LED which lights up at the counter
output and so it will be clear who has
to dip their hand into their pocket in
order to pay for the next round.

If the counter happens to stop at 'zero',
no LED will light and then the round
will have to be paid for out of the 'kitty',
unless of course the landlord happens
to be in a generous mood! If the honour
of buying a round should happen to fall
on an absentee (whose switch will be
in the B position) the monostable will
be retriggered so that the counter
continues to count until one of the
people present is selected.

Figure 3 shows a suitable printed circuit
board for the project. In comparison
with the low constructional costs
involved, the performance of the drinks
round indicator is extraordinarily high,
provided, of course, the constructor
is a member of the 'club' and made
exempt of all payment. N

Editorial note: No responsibility will
be accepted for debts and/or medical
expenses incurred by repeated use of
the rounds indicator. Depending on the
licensing laws in your area, a single
battery should not run out before
closing time, but it may be an idea to
carry a spare . . .

In this particular article a certain
amount of mechanical skill is required
with electronics taking a back seat for a
change. From the diagram in figure 1 it
will be seen that, as far as the electronics
is concerned, one 4093 and two 4026
IC's and five resistors are all that is
required.

The gates N1 and N3 form two very;
simple oscillators. When the start
button S is pressed, these generate
the clock signals for the two counter-;
decoders, IC1 and IC2. The outputs]
of the decoders are fed directly
total of 15 red, yellow and green LEDs
which are combined to form a multi-
coloured triangular display. The clock
frequencies of the oscillator are low
to give the impression of random ti
of coloured trails within the triangle.

12-22 - elektor december 1980 kaleidoscope

3

O yellow

Figure 3. This suggested layout for the LEDs
has proved to be very effective in practice.

A further possibility is the use of a plug and
socket arrangement between the LEDs and
the circuit board. This would allow a change
of pattern.

The display is even more attractive
when three 'mirrors' are placed to form
a triangular 'tube' as in the manner of a
normal kaleidoscope. The reflecting
sheet in the prototype was in fact three
pieces of polished copper clad board.
However, it would not be unreasonable
to find handbag mirrors that would fit,
the improved results would certainly
be worth the effort.

The construction is illustrated in figure 2
and is relatively simple. Two cans are
used to give sufficient depth to the
completed container. The upper can
needs to be long enough to contain the
mirrors, the display board and the
electronics boards. The lower section
will then suffice as the battery holder
and cap. Some experimenting with
different ideas will be useful for joining
the two can halves together, see the
article on can opening.

The three 'mirrors' can be epoxied
together to form a triangular tube
which is then mounted on a disc of
perspex used for the LED display. The
electronics, discs of Veroboard, and a
protective cover of perspex, can the be
fitted using threaded spacers. The lower
can section, the 'cap', might then poss-
ibly be fitted using a central threaded
spacer. This should leave ample room
for one (or two in parallel) 9 volt
batteries standing vertically.

It should be emphasised that the cut
edges of the can need to be carefully
protected since the finished kaleido-
scope is likely to encourage a lot of
interest, particularly among children. M

Obviously, it would be unwise to
expect sensational technical data and
precision from such an instrument.
However, as illustrated in figure 1,
the panoramascope is undoubtedly
original and very easy to operate. When
the device is switched on, the 'drum'
containing the row of LEDs will rotate
directly in front of the observer. If a
DC or AC signal is then connected to
the Y input, the row of LEDs will light
in sympathy with the signal, and an
oscilloscopic image will be built up.
This proves that a mechanical X deflec-
tion can successfully be employed to
display an alternating (or DC) signal
on a 'screen'.

In actual fact, it is an optical illusion.
When light spots are rotated at a high
an oscilloscope with s P eed our relatively slow optical sensors

u | */ * ri - are no longer able to distinguish be-

mechamcal X deflection tween the sections of the picture.

panoramascope

After reading the subheading, the serious minded electronics
enthusiast may be excused for thinking that Elektor have taken a step
backwards in circuit design. Why do we need to use anything
mechanical in oscilloscope construction (apart from the knobs and
switches) with all this amazing 'micro chip' technology we know
about? Well, we asked for original and innovative ideas and we
certainly got them. Actually, this device is not as peculiar as it may
appear at first sight.

The inventor of this particular piece of electronic wizardry has made
full use of the geometric properties of a tin can and has combined
them with a drop of mechanics and a spot of electronics to produce
an oscilloscope. The end result is a panoramascope which in principle
dates back to the days when pictures first started to move.

H. Luhmer The mechanics involved

Figure 1 illustrates the (not insurmount-
able) mechanical problems that face
the enthusiast wishing to construct
the panoramascope. A slot is cut into
the side of the can behind which the
row of LEDs are mounted. Holes are
drilled in the centre of the top and
bottom of the can where the 'bearings'
are to be mounted. The author actually
used jack plugs and sockets. In this
way the supply voltage can be fed in
through the top plug and socket com-
bination while the (Y) input signal
to the display board can be fed up
through the lower pair. It is not certain
for how long these bearings will stand
up to the mechanical stresses and
strains of everyday rotation, but no
doubt the inventive enthusiast will
devise a more elegant method of solving
the problem. Possibly by feeding the
supply voltage and signal into the unit
by means of 'slip rings'.

In order to spread the forces acting on
the can during rotation as evenly as
possible, a threaded rod has been
positioned diametrically opposite the
central axis of the board and a counter-
weight has been attached to the wall

of the can directly opposite the row
of LEDs. The exact weight and pos-
itioning of the latter may require a
little experimentation for optimum

The drive motor and its associated
electronics are housed in the 'control
box' situated at the base of the panor-
amascope. The recommended motor
is a 'mini drill' type that can run on
batteries. It is also necessary to be able
to control the speed of the motor. A
simple and suitable motor speed control
circuit which utilised an adjustable
voltage regulator 1C, appeared in Elektor
number 57 (January 1980, page 1-16).
Alternatively, an adjustable power
supply can be used to control the
speed of the motor.

The can is 'hung' upside down so that
the jack plugs on the base of the control
unit and on the retaining arm, which is
firmly bolted to the unit, fit neatly
into the sockets on the can. It is very
important that the can is able to rotate
on its bearings with the minimum of
effort. A pulley is then placed on the
shaft of the motor and a rubber belt
placed around it and the groove in the
top (now on the bottom) of the can.
When everything has been correctly
aligned the pulley can be fastened onto
the shaft.

The potentiometer for the adjustable
power supply is the only manual control
required (apart from the motor switch).
This is mounted on the front panel of
the control box as it constitutes the

elektor december 1980 - 12-23

'time base' control. The amount of X
deflection is controlled by the DC
motor, thus it is not so mechanical after
all!

The electronics involved

As can be seen from the circuit diagram
of figure 2, there is very little involved
in the way of electronics. The circuit
consists of two ICs, a few odd com-
ponents and 16 LEDs, all of which
will fit very neatly on a small piece of
circuit board.

The 16 LEDs are mounted in a row to
form the Y axis of the panoramascope.
Unfortunately, the device can only
be used to measure positive DC signals
as IC1 will not withstand negative
voltages! The maximum DC input is
+5 V and the zener diode D17 has been
included to protect the circuit from
input voltages that exceed this level.

The voltage divider network comprised
of R2, R3 and PI sets the bias voltage
at the input of IC1 to about 2.5 V. This
means that an AC signal can be fed in
via Cl as long as the signal remains
within the limits imposed by IC1
(2.5 V pp = 1.77 V eff ).

The circuit can be battery powered (two
9 V batteries in series) or a transformer
and bridge rectifier can be added. In
either case, IC2 stabilises the supply
voltage to +12 V.

Once the panoramascope has been
constructed, the next step is to test it.
Potentiometer PI should be adjusted
so that one of the centre LEDs is
illuminated. An AC signal, with an
amplitude no greater than 1 .77 V e ff,
is then applied to the Y input socket.
The motor is then switched on and the
'picture', which will inevitably be
moving, is synchronised with the aid of
the time base potentiometer until it
appears stable.

Note that, as with all oscilloscopes, a
DC and AC signal should not be applied
at the same time. If desired, a switch
can be incorporated to select between
the two.

This machine is guaranteed to produce
'spots in front of your eyes!' M

Figure 2. The circuit diagram of the panoramascope. The unit can be powered either by batteries or by (rectified and smoothed) mains.

Yes it is a locomotive, no it will not run
and no we do not know the prototype!
However, it is original and well within
the spirit of the canned circuits com-
petition. We did not actually say that
the circuits must work did we? This is,
in fact, the only entry that has all the
electronic components outside the can
- so we invented a special class for it —
and a prize!

According to its creator, only items I
from the junk box were used. Hardly
any holes were drilled or screws used.
The main item in the tool kit consisted )
of a 50 watt soldering iron and (suf-
ficient) solder.

Ireclir isliims

HI I |)< k tilioil

The prize

Just to be different, we have decided
to make the locomotive the basis of a
competition. A prize of a bedside radio
(an unusual one at that) will be won by
the first reader to send in the most
complete and correct list of all of the
components used in the construction
of the locomotive. We do not expect
a 100% correct entry so give it a try.
Most, if not all, of the components
will be familiar to many but the identity
of a few parts are reminiscent of the
valve days and will possibly elude our
younger readers, although we could be
mistaken about that. It's fun and it
could take quite a time identifying all
of the parts - if you can. The can? Oh
yes, that's the boiler. K

locomotive

It doesn't squeak, play tunes, light up or move about. But it does
include a can, electronic components and it took quite a lot of
imagination to work it all out, not to mention mechanical expertise
to build it.

Competition details

Mark the envelope 'Loco' and post the
completed entry to reach the Elektor I
Canterbury office by the first post on I
the 5th January 1981 . The judges I
descision must be final and no corre-
spondence can be entered into.

44

cumulalhe Index

analogue delay technology 2-34

FET op-amps in the FORMANT 2-09

frequency doubler 1-22

1C tremolo 7-62

legato gate interface for FORMANT 7-15

loudspeaker fuse 7-23

loudspeaker fuse 7-78

moving coil preamp 7-98

PWM amplifier 5-22

stereo dynamic preamplifier 7-05

super tiny amplifier 7-65

talk funny on board 1-30

ulp amp 7-40

V-FET amplifier 7-10

voltage controlled harmonic generator 7-67

Car and bicycle

active car aerial 4-24

afterburner 4-45

aide-de-camp 4-31

automatic battery charger 4-44

automatic cycle lighting 7-07

battery protection 4-08

battery voltage indicator 4-30

car of the eighties 4-03

car stabiliser 7-16

dip-stick probe 4-16

fuel consumption meter 4-34

improved anti-theft device 7-38

intelligent wiper delay 4-18

power flasher 4-46

r.p.m. indicator as an economy guide 11-04

security rear light 7-38

simple fuel economy meter 11-16

simple wiper delay 4-43

solid state tachometer 8-00

stop thief I 4-28

thyristor ignition system 2-32

transistor ignition 4-10

Domestic and hobby

aide-de-camp 4-32

automatic curtain control 11-30

automatic pump control 11-25

bath thermometer 1 2-07

cheap light meter 7-14

coffee machine switch 11 -42

current saving home trainer 7-24

digital thermometer 2-44

draught detector 11-08

dual slide fader (1) 3-24

dual slide fader (2) 10-36

electronic garden gnome 12-48

energy monitor 7-59

energy saving motor control 11-39

exposure meter and development timer 7-08

flexible intercom 5-44

fridge alarm 11-34

garden path lighting 7-14

infrared receiver 7-02

infrared transmitter 7-02

midnight raid detector 12-12

morse trainer 6-35

operational hours counter 11-44

post indicator 7-74

private telephone exchange 7-60

programmable slide fader 10-05

remote control 3-36

remote control dimmer 7-35

remote control slide projector 10-24

sensibell 7-04

smoke detector 6-32

temperature transmitter 7-34

Timbug II 6-18

touch doorbell 10-10

TV ... interference suppression 2-12

VOX printed circuit board 9-22

water detector 7-22

wet alert 7-25

Fun, games and model building

a flash in the can 12-16

armless bandit 7-64

can shy 12-10

dice shaker 12-29

din can 12-39

drinks round indicator 12-19

enigma 7-26

genie in a can 12-40

kaliedoscope 12-20

locomotive 1 2-24

midnight raid detector 12-12

missile attack 7-69

mobile can 12-38

multigame can 12-35

musical box 7-30

musical cube 6-20

sensibell 7-04

Tilt 12-33

Timbug II 6-20

water can 12-46

Generators

colour generator 1-06

crystal controlled sinewave generator 7-81

digital-sinewave generator 7-16

noise at high frequencies 6-34

precision VCO 7-40

sound effects generator 7-58

variable pulse width generator 7-74

video pattern generator 1 0-28

voltage controlled harmonic generator 7-67

HF

active car aerial 4-24

aerial amplifiers 2-27

aerial booster 2-04

AFC via the tuning diode 7-12

alternative amplifier for the J2-aerial 7-12

deglitching remote control 7-88

digital tuning scale 1-08

frequency lock system 5-18

FSK PLL demodulator 7-87

high frequency optocoupler 7-37

LCD tuning scale 10-32

low-pass filter 7-71

narrowband IF strip 6-22

noise at high frequencies 6-34

remote control 3-36

remote control dimmer 7-35

remote control slide projector 10-24

selective CW filter 7-69

simple remote control 12-44

state variable filter 7-52

super regenerative receiver 87-180 MHz 7-80

test transmitter 7-19

TV . . . interference suppression 2-12

variable HF oscillator 7-39

VSWR meter 1-18

O aerial 6-08

4.4 MHz crystal filter 7-06

Informative articles

analogue delay technology 2-34

candid comments 12-17

car of the eighties 4-03

eighties, the 1-04

ejektor, measuring multipath 9-26

electrolytology 9-33

elektor jiP's 504

energy saving know-how 11-13

energy saving motor control 1 1-39

how to recycle dry cell batteries 11-10

1C guide 7-49

Josephson computer 9-15

know the ins and outs of your central heating .. 11-36

LCDisplays 5-34

long life technique in light bulbs 11-28

modern methods of voltage regulation 1-32

more on TV games 6-26

more TV games 10-17

PWM applications 1-36

switched capacitors 10-12

TV . . . interference suppression 2-12

using the vocoder 9-38

Microprocessors

BASIC cassette interface 5-24

cassette interface adaptation 7-09

elekterminal width extender 6-38

elektor pP's 5-04

FSK PLL demodulator 7-87

hexadecimal display decoder 7-83

hexadecimal keyboard 7-72

high speed readout for elekterminal 9-28

Junior Computer 5-08

Junior Computer memory card 1 0-22

more on TV games 6-28

more scope for the elekterminal 7-22

more TV games 10-17

new busboard for microprocessors 1-31

printer for microprocessors 3-28

PROM programmer 7-90

RAM tester 7-70

RS 232 interface 7-77

RS 232 line indicator 7-30

SC/MP as a mini organ 6-37

small pP switching supply 7-72

synchronous FSK 7-17

8 k RAM + 4, 8 or 16 k EPROM on a single card 9-04

Miscellaneous

canine defence 1 2-47

cheap seconds 7-71

cheap source of energy 7-99

disco lights controller 6-06

electronic fuse 7-34

EXOR edge detector 7-29

frequency conversion 7-02

frequency lock system 5-18

how does this strike you? 7-95

hybrid cascode 7-16

knock switch 7-68

odd divider 7-57

pest pester 6-04

plus or minus unity gain 7-25

PWM battery charger 7-18

running lights 7-76

sawtooth synchronous to mains 7-26

shop window lighting 7-92

skin resistance biofeedback 7-29

trigger with presettable thresholds 7-42

universal display 7-84

universal warning alarm 7-28

variable logic gate 6-30

voltage controlled duty cycle 7-08

voltage drop detector 7-77

Music

better piano 7-85

chorosynth 3-17

clavitar 3-04

FET opamps in the FORMANT 2-09

legato gate interface for FORMANT 7-15

marimba can 12-34

musical box 9-30

piano that sounds like a piano 3-13

12-28 -ele

piping can 12-42

sample and hold for synthesisers 7-42

talk funny on board 1-30

two more piano octaves 7-89

using the vocoder 9-38

vocoder ( 1 ) 1 -24

vocoder (2) 2-14

xylophone 1 2-09

Power supplies

aide-de-camp 4-32

converter 7-06

intelligent NiCad charger 7-22

NiCad piggy bank 12-13

precision power unit 9-08

simple symmetrical supply 7-75

stabilised 10... 350 volt power supply 7-79

variable power supply 0-50 V/0-2 A 7-86

2 amp voltage regulator 7-03

-12 V from +5 V 7-34

Test and measuring equipment

acoustic ohmmeter 7-63

audio frequency meter 7-88

automatic range switch 7-99

battery voltage indicator 4-30

canometer 12-30

continuity checker 7.95

digisplay 2-32

digital heart beat monitor 7-96

digital tuning scale 1-08

dip-stick probe 4-16

electronic linear thermometer 9-12

exposure meter and development timer 7-08

extending the 1/4 GHz counter 2-14

fuel consumption meter 4-32

luxury transistor tester 6-14

mini speed control 1-16

NiCad charger 7-37

panorama scope 1 2-22

phase meter 7-27

RAM tester 7-70

RMS to DC converter 7-75

RS 732 line indicator 7-30

semiconductor barometer 7-33

simple analogue voltmeter 7-30

simple L/C meter 7-44

simple opamp tester 7-07

simple 555 tester 7-31

test transmitter 7-19

VSWR meter 1-18

wind-o-meter 7-13

link SO

an imktv to missing links

The purpose of the Link is to assist the home construc-
tor by listing corrections and improvements to Elektor
circuits in one easy to find place. At a glance it can be
seen whether a project involved any problems.

Don't forget to check previous Links if the project in
question was published before January 1980.

E60

ICU — a mini microprocessor
March 79IE47I. p.3-28
Burglar’s battery saver
Jul./Aug. 79 (E 51/521, p. 7-05
Pcb for variable fuzz box
Oct. 79 IE 541. p. 10-20
Charging NiCads
Dec. 79 (E56). p. 12-10

Dec. 7?(E56), p. 1 2-10
Colour generator
Jan. 80 (E 57) p. 1-06
Vocoder

Jan./Feb. 80IE57/58). p. 1-29. p. 2-16
ESS 005 Luna

E61

Car collision alarm
Jul./Aug. 79 (E51/52), p. 7-89
Top Preamp
Dec. 79IE56). p.12-15
Chorosynth

March 80 (E 59). p. 3-17

E62

Top Preamp
Dec. 79 (E56).p. 12-13
Elek terminal
Active car aerial
April 80 (E60). p. 4-24

E66

8 k RAM card
Sep. 80 (E65). p. 9-04

E67

NiCad charger

Jul./Aug. 80 (E 63/64). p. 7-37
VFET amplifier
Jul./Aug. 80 IE63/64). p. 7-10
Car alarm

April 80 IE60), p. 4-28

E68

Programmable slide fader
Oct. 80IE66), p. 10-05

- 12-29

dice shaker

R. Mohunlol

eiektor december 1980

Not an ice shaker, but a dice shaker, an
ingeniously simple idea. No high tech-
nology electronics involved, but then
that was hardly the object of the
circuits in this issue.

What's it all about? Well, throwing dice,
only this time using an electro-mechan-
ical design based on the ordinary,
old-fashioned dice.

The constructional drawing in figure 2
illustrates the principle. The dice are
placed on the diaphragm of a small
loudspeaker and are shaken by feeding
the loudspeaker with a 50 Hz signal
derived directly from the secondary

Figure 1. In the circuit of the dice shaker, the switch SI is a double pole pushbutton. A -biased - toggle switch would also be suitable.

winding or a small transformer
(9 . . . 10V/0.5 A).

The basic minimum of components
required are a can, a loudspeaker, a
transformer and a switch. To be on the
safe side, however, it is advisable to
include a timer controlled by a random
pulse width generator, to make absol-
utely sure that the length of time for
which the dice are shaken is unpredict-
able.

Figure 1 shows the circuit of the shaker.
The loudspeaker is connected across the
secondary winding of the transformer
via the contacts of the relay controlled
by the 555 timer IC2. Pressing the start
button SI will trigger IC2 (SI a), and a
100 Hz pulse train will be fed from the
bridge rectifier, via Sib, to IC1. This
4 bit counter selects a random RC time
constant for IC2 from the networks
R1/C1, R2/C1 or R4/C1 and thereby
determines the time during which the
relay is switched, in other words, the
'shaking time' for the dice.

If an 8 ft loudspeaker is used, a 12 £2

3 watt resistor must be connected in
series with it. The value of the resistor
must be raised to 16 fi/4 watt with a

4 S2 loudspeaker. It is preferable to use
a relay with a high coil impedance,
certainly not less then 300 S2.

It must be emphasised that the mains
earth lead must be a good secure con-
nection to the can. M

12-30 - 1

Distances can be measured in three
basic ways : by using a ruler, a tape
measure, or by electronic means. The
first two are often time consuming and
prone to inaccuracy and the third is
relatively expensive. A fourth possi-
bility is now provided by none other
than this canometer, its prime merit

an electronic meter
an empty can and
indicates the measured distance
digital display.

The principle behind the circuit could
hardly be more basic. When a cylinder
is rolled along a flat surface, the distance
that it covers during one revolution will
be the same as its circumference. If
of the cylinder, or can.

measuring can

S. Heilmann

The fact that a can has
volutational properties was the
inspiration behind this particular
design in which the roll plays a
very important role. The circum-
ference of the can is used to
measure distances with a fair
degree of accuracy. The distance
covered, or rather, rolled is
indicated on a four digit display,
the maximum readout being
99.99 metres.

is known, along with the number of
revolutions the can has made, the dis-
tance covered can be calculated very
easily. Therefore, the electronics in-
volved only has to record the number
of revolutions, calculate the distance
travelled and display the result. How-
ever, with a little bit of ingenuity, we
can dispense with the mathematical
section altogether, thereby making the
electronics even simpler.

The diameter of the common 0.33 litre
can is approximately 65.8 mm. This can
be increased to 66.8 mm by adding a
layer of sellotape around the can. This
will then bring the circumference of the
can up to 21 cm, a nice whole number.
To detect how far it has travelled, the
can utilises a disc which has 21 segments.
Each segment consists of a dark and a
transparent section. The disc can
revolve on a central axis and is weighted
with either lead or solder so that the
weighted portion will remain face
downwards when the can is rolling. A
photo detector construction from an
LED and a photo transistor, on either
side of the disc, is incorporated so that
the transistor generates a pulse each
time a segment rolls past. Since there
are 21 segments and the can has a
circumference of 21 cm, you don't need
a degree in mathematics to work out
that each pulse from the photo transis-
tor corresponds to 1 cm of travel!
Effectively, the can rolls together with
all its contents (circuit boards, battery
etc.) but the disc stays still.

Circuit diagram

The circuit diagram of the canometer
is shown in figure 1 . The photo detector
mentioned previously is made up from
LED D2 and the photo transistor T2.
Whenever the light from the LED passes
through transparent section of the disc
transistor T2 will generate a pulse. This
pulse is shaped by the schmitt trigger
N1 and fed to the clock input of IC2

via gate N2. However, the pulse will
only be allowed through N2 when the
output of the flipflop, formed by N3
and N4, is high. In other words, after
the start button, S2, has been depressed.
If the stop button, S3, is pressed the
clock pulse will be inhibited again and
any reading obtained will remain on
the display. The counter section of IC2
will therefore only operate after the
start key has been depressed and the can
is rolling. Switch S4 serves to reset the
circuit.

All the functions required to process
the pulse data (counter, decoder and the
multiplex control circuitry for the
displays) are contained in IC2. Common
cathode displays are used and the
decimal point of Dp2 is utilised to
distinguish between metres and centi-

Although IC2 requires a regulated
supply voltage of 5 V, it appears to
work quite happily at lower voltages
as well. A supply voltage slightly greater
than 4 V was therefore selected so that
three 1.5 V batteries (connected in
series) can be used for the power source.
A 9 V battery could also be used if
the canometer is not going to be rolled
very often. To keep dissipation down to
a minimum a germanium transistor (T1)
has been incorporated to stabilise the
supply voltage. The circuit has only one
adjustment point, potentiometer PI.
This regulates the sensitivity of the
photo transistor.

Construction

The complete circuit can be mounted
on two printed circuit boards as shown
in figures 2 and 3. The round display
board is mounted at right angles to the
rectangular board after which the
corresponding connections are made
with short wire links.

The axle for the segmented disc is
mounted on the rectangular board in
the position marked by a dotted line.
The space with the triangular points
next to it has to be removed first with
the aid of a fretsaw. This is essential so
that the disc will be positioned correctly
between the LED and the photo transis-

The segmented disc is illustrated in
figure 5. In fact this can be carefully
cut out and glued onto a circular piece
of perspex with the same diameter. This
is then weighted with lead or solder,
bearing in mind that the completed
disc must be able to pass through the
slot in the board without touching the
sides. Once it is mounted on the axle
it must be able to turn freely.

Prior to final assembly in the can, preset
potentiometer PI should be adjusted so
that the display increments by one
each time a segment of the disc passes
through the light beam (after S2 has
been depressed of course!).

The only other requirements are that
the can needs to be 'screened' so that no
light can enter to cause a misreading and

00000000

the axle for the segmented disc must be
positioned in line with the central
axis of the can.

Operation

The directions for use as given by the
author are so clear there is no need to
add anything to them:

* Switch on the canometer

' Place the can at the starting point of
the distance to be measured, press
'reset' followed by 'start'

* Roll the can as evenly as possible
until the 'finish', then press 'stop',

pick up the can and read off the dis-

tance measured from the display

* Note that the unit will only work
correctly when the can is rolled in

one direction. If the canometer is put
into reverse half way through a measure-
ment an incorrect reading will result.

* Interim measurements are also poss-
ible: 'stop' is pressed at the required

point, the distance indicated on the
display is noted and the can is replaced
in the same position. Then press the
'start' button and continue.

* Do not forget to switch off the
canometer after use.

* Roll on! M

elektor december 1980 — 12-33

Remember the old 'ring the bell' game
that consisted of a bent length of wire
and a loop on a handle, where the
object of the exercise was to see who
had the steadiest hand? Elektor now
present a very compact version of the
game which simply involves the moving
of a can from one place to another
without it making a (somewhat ter-
rifying) noise. The basic circuit could
also be employed in many other appli-
cations such as an electronic plump-bob
or level indicator (audio spirit level?).

In this particular instance the cylindrical
shape of the can has been used to its
fullest advantage. The most important
component being a pendulum which

P. Groger

reacts to any movement of the can by
swinging and thereby effectively closing
a switch. As can be seen from the block
diagram, the remainder of the circuit
is very simple. The pendulum triggers
a monostable multivibrator which in
turn activates an astable multivibrator in
the form of an audio tone generator.

As soon as the pendulum touches the
contact loop (see circuit diagram), the
electrolytic capacitor at the input of
the circuit will charge up rapidly via
R2. Transistor T1 will conduct and
activate the Schmitt trigger formed by
T2 and T3. which turns on the oscillator
(T5 and T6) via transistor T4.

If the can is at rest and the pendulum
no longer touches the contact loop,
the electrolytic capacitor. Cl, will
discharge through the base-emitter junc-
tion of T1. Only when the capacitor is
discharged to below about 0.7 V will T1
turn off thereby inhibiting the oscillator
via the Schmitt trigger and T4. This
means that the oscillator will continue
to produce a noise for about 5 seconds
after the can has 'settled down’. If a
longer duration is required, the value
of R2 can be decreased and the value
of Cl increased.

The 'sensitivity' of the instrument
depends on the diameter of the contact
loop and its distance from the weight
of the pendulum. The smaller the
contact loop and the closer it is to the

weight, the more sensitive to movement
the device will be.

The circuit has high operational re-
liability, low quiescent current con-
sumption and is economical to con-
struct. Being camouflaged, the circuit
has another application as a burglar
alarm. Who would expect a beer can to
start making a noise? H

marimba can

The steel band plays a significant part in the characteristic sound of
Caribbean music. The mainstay of this sound is the marimba, a steel
drum with different sized indentations in the top. It is played in a
manner that is a cross between the drums and the xylophone. The
originals were, in fact, constructed out of old 50 gallon steel oil drums.

Although beer cans are made in a
similar manner to oil drums, they
certainly do not produce anything like
the same noise. Of course, size might
have something to do with it (maybe
50 gallon beer cans would be quite a
good idea) and in this article we use
electronics to compensate for this
deficiency. The playing technique re-
mains unchanged, a drum stick can still
be used but with the addition of a small
magnet fitted to its end. The top of
the can hides a reed relay and it is
this that controls the electronic sound
producing circuit inside the can.

Each can produces a single tone, so the
more cans used, the wider the range

of tones available. Since it is possible
to use any 'make' of can, a selection of
different ones could be useful to make
it easier to remember which can con-

As figure 1 shows, a number of cans
need not be expensive even when
taking into consideration the cost of
the full cans in the first place. Anyway,
think of the excuse we offer you,
'I am just going out to get some more
cans for my marimba set'.

The CMOS 4011 1C contains four
NAND gates, two of which (N 1 and N2)
provide an astable multivibrator to
generate the tone. The other two gates
(N3 and N4) are used to provide the
base drive current for transistor T1
which, in turn, drives the loudspeaker.
The can and the loudspeaker together
act as a resonant cavity to produce
the characteristic sound of the marimba.
The RC time constant of the astable
multivibrator determines the pitch and,
using the values given (C = 22 n,
R=100k) will be about 150Hz. If
the 100 k resistor is replaced by a preset
potentiometer of the same value, it
will be possible to tune the can comfort-
ably within several octaves from the
desired pitch.

Battery consumption will be very low
since the reed switch only operates for
a very brief period. In other words,
battery life is dependant on how often
you play. Once a night should be
enough. H

ir december 1980 - 12-35

Nine assorted switches and a potentio-
meter provide so many variations that
one and the same circuit can be used to
play at least six different games:
throwing dice, heads-or-tails, snakes and
ladders (versus an opponent or against
the on board 'computer' if desired) and
two variations of the well known
'skeet'.

As can be seen from the diagram in
figure 1, the heart of the circuit is a ten
channel running light: IC1 acts as a
clock generator for the divide-by-ten
counter IC2 which controls up to ten
LEDs directly. The remainder of the
circuit consists of a flipflop constructed
around N1 and N2, an oscillator formed
by N4 and a loudspeaker amplifier
stage, T2.

To describe the complete operation of
the circuit it is best to examine the
operations involved in each game in

multi-game can

This can is so versatile that a
number of different games can be
played with it and it can also be
used as an egg timer. All it
involves is a series of switches and
some electronics to alter the
characteristics of a simple running
light display. The circuit also
includes a modest sound effects
unit to provide extra dimensions
to the game being played. All the
electronics involved can be
squeezed onto a printed circuit
board that will fit inside a
'standard' can.

R. de Boer

Throwing dice

For this game the rotary switch, S3,
is placed in position 6, S2 is closed, S4
is placed in position (a) (reset), S5 is
left open, S8 is placed in position (a)
and the setting of potentiometer PI is
adjusted to give an output frequency
of 20 ... 30 Hz.

If switch SI is then closed briefly the
die will start to roll. After a preset roll
time, adjustable by means of potentio-
meter P2, the result will be shown by
one of the first six LEDs lighting
(D10...D15).

Operation

Since the rotary switch, S3, is in pos-
ition 6, the counter will always be reset
when it reaches this number (via T3 and
S4). Providing a clock signal is present
therefore, IC2 will keep counting from
0. . .5.

When SI is closed, capacitor 01 will
charge up rapidly via R1. Transistor
T1 will then start to conduct, activating
the clock generator IC1. This if course
starts the die (in fact a six channel
running light) rolling at a speed deter-
mined by the setting of PI. When SI
is released, capacitor Cl will slowly
discharge through P2 and R5. There-
fore, transistor T1 will gradually turn
off and the die will slow down before
coming to a halt. The period of time
that this takes is determined by the
setting of P2.

The roll of the die is rendered audible
by feeding the clock pulses generated
by IC1 to the loudspeaker interface
via diode D26.

Heads-or-tails

Now that we have discovered how to
throw dice with this circuit, the follow-
ing game will be as easy as falling off a
log. The operation is virtually identical.
The only difference being that S3 is

now moved to position 2. This means
of course that the counter can only
count to two and only two LEDs can
light: D10 for 'heads' and Dll for
'tails' — or vice versa.

Snakes and ladders (1)

For this game the positions of the
switches have to be altered slightly: S2
and S5 will be 'off, S3 in position 9,
S4 in position (b) (stop) and S8 in
position (a). Potentiometer PI should
be adjusted to give a frequency of
around 1 Hz.

During this game, two players may
alternately 'climb' one, two or three
rungs of a 'ladder' (the ladder is the row
of LEDs). This is done by keeping SI
closed until the LEDs indicate the
number of rungs that the player has
climbed (the number of rungs to be
climbed must be stated before the
player presses SI). The player to cause
the last LED to light is the winner and
this is proclaimed by a squeaking tone
from the loudspeaker.

Operation

Again a running light display is used,
only this time at a much lower fre-
quency. Switch SI starts the oscillator
and causes the LEDs in the row to light
one after the other. When output 9
of IC2 becomes high the last LED in
the row will light and, via S3, T3 and
S4, the clock enable input of the coun-
ter will also go high. This causes the
counter to stop. At the same time, the
oscillator formed by N4 is enabled via
D28. The output of this oscillator
is made audible via D27 and the loud-
speaker amplifier T2. The on/off
switch, S9, can be used to reset the
circuit.

Snakes and ladders (2)

This is virtually the same game as
before, but this time the on board
'computer' is the opponent. Whenever
it is the 'computer's' turn, S8 is
switched to position (bf for a short
period to obtain its 'opinion'. One of
three LEDs will light to indicate
whether it wishes to climb one, two or
three rungs of the ladder. When S8 is
moved back into position (a), switch SI
is operated to carry out the 'computer's'
request. After that it is the turn of the
human once more.

Although the computer isn't up to
much electronically speaking, you will
soon find out that it can be a formidable
opponent to beat! It is guaranteed to
win in any game that it is allowed to
start (every other game) provided there
is no cheating!!

Operation

Diodes D1 . . . D9, together with S8,
constitute the 'computer'. When S8 is
switched to position (b) one of the
LEDs Dll, D12 or D13 will light to
indicate that the computer wishes to
ascend one, two or three rungs respect-
ively. In this position, via S81, the

12-36 - elektor decer

multi-game can

Figure 1. The electronic interior of the multi-game can. It all centres around the running light constructed around IC1 and IC2.

'ladder' LEO is suppressed to avoid
confusion.

Skeet (1)

This game requires yet another combi-
nation of the switches. This time place
SI to the 'on' position, S3 to zero,
S5 to 'off' and S8 to position (a). The
speed of the game can be adjusted by
means of PI as required (between about
3... 30 Hz).

The object of the game is to 'shoot'
the light running from left to right on
the row of LEDs. When the sixth LED
in the row lights (D15), either S6 or
S7 must be depressed very quickly. If
one of these switches is depressed at
exactly the right moment, a buzz will
be heard and the target will remain
stationary for a split second after which
it will continue on its journey.

As a result of depressing either S6 or S7
a pulse is generated which will have
absolutely no effect provided that
output 5 (pin 1) of IC2 is low. If this
output was high, however, the pulse
will be allowed to pass. The output
of N1 or N2 (depending on whether it
was S6 or S7 that was depressed) will
then go low for about 1 Vi to 2 seconds
(determined by the networks C5/R10
| and C6/R1 1 ). During this time the clock

generator is inhibited via D22 or D23 so
that the target is stopped.

If the output of N1 goes low, the
output of N3 will go high for about
half a second (determined by C8/R4).
If on the other hand, it was the output
of N2 that went low, the duration of
the pulse at the output of N3 will be
set to about one second by C9 and R 18.
In either case, the oscillator (N4) will
be enabled to accompany each 'hit'
with a buzzing tone - a short one if
S6 was pressed and a longer one if S7
was depressed.

Skeet (2)

Again, the target is shot as it moves
from left to right across the display, but
now two players are involved. This gives
rise to a great deal more excitement.
Both players have a separate button,
either S6 or S7. The object of the
exercise is to see who is the fastest
shot. This will be indisputably deter-
mined by the duration of the buzz — as
explained above. (Note: the switches
must be released immediately otherwise
there will be no audible difference
between the length of the tones pro-
duced by S6 and S7).

Operation

The flipflop, N1/N2, removes any
doubts concerning which player has the

fastest reactions. As soon as one of the
pushbuttons is depressed the output
of either N1 or N2 will go low and
inhibit the other gate. Thus, the other
player's shot will have no effect
whatsoever. If both players fire too
early, or too late, the target will simply

Egg timer

As mentioned earlier, the multi-game
can can also be used as an egg timer. For
this application switches SI and S5 are
placed in the 'on' position, S2 to 'off',
S4 to (b) (stop), S8 to (a) and S3 and
PI are set as required.

This particular egg timer again utilises
the running light display (did you
expect anything else?), but this time
of course it operates at a very, very low
frequency. In fact, more of a crawling
light!! With S5 closed an extra large
capacitor (C2) is included in the fre-
quency determining network of the
clock generator. According to the
setting of PI it can now take between
1 54 ... 80 seconds for the 'snail pace'
light to move from one position to the
next. Absolutely ideai for an egg timer,
or any other kind of timer for that
matter. If, for instance, PI is set to give
a pulse every 30 seconds and S3 is
switched to position 6 an alarm tone
will sound after 3 minutes (6 x 30

Resistors:

R1.R7.R8 = 10 k

R2.R15 ■ 8k2

R3.R22- 100 k

R4.R6.R 1 7,R18,R24 = 1 M

R5.R19 = 1 k

R9 * 1k5

R10.R11 =■ 1M5

R12- 150J1

R13.R14 " 15k

R16 = 150k

R20.R21 ,R23 = 12 k

PI - 470 k potentiometer

P2 ■ 470 k preset potentiometer

Cl =■ 22 ji/16 V
C2 ■ 220 ji/16V
C3 = 4p7/16V tents
C4 - 47 n
C5.C6.C7 = 270 n

C7 = 150 n
C8 = 92 n
CIO = 22 n
C11 = lOn
C12 » 10 p/16 V

Semiconductors:

T1.T3- TUN
T2 = BC510

D1 . . . D9.D20. . . D28 = DUS
DIO . . . D19- LED red
IC1 = 555
IC2- 401 7 B
IC3 - 4093

Mi

S1.S2.S5.S9 * SPST sv

53 * 10 position rotan,

54 = SPDT switch
S6.S7 = pushbutton
S8 * 4 pole double thri
LS - 0.2 W/8 Si loudsp

seconds). All sorts of times can be
preset in this manner - within reason
of course!!

Construction

There is no constructional drawing
for this particular project, but it should
not be too difficult to manage without.
Unlike some of the other circuits
published in this issue, very little
mechanical knowledge is required to
construct it. The loudspeaker is best
mounted on the base of the can. The
ten LEDs should be mounted in a neat
row so that they are visible from the
outside of the can - either on the side
or on top underneath a perspex cover.
All the switches as well as the potentio-
meter PI should be accessible from the
outside. The remainder should not
cause any problems, because a printed
circuit board has been designed for the
complete circuit (see figure 2). This
board is small enough to fit inside a
normal can. The supply may be pro-
vided by a single 9 V battery. Being
so small, it can easily be squeezed in
somewhere. M

12-38 -

• decamber 1980

mobile i

mobile

can

R. Wenzelburger

the true 'tin robot'

Although no great claims are being
made as to the practical
applications of this circuit, the
final result is both interesting and
amusing. The circuit also provides
a peek (albeit a very small one!) at
the amazing sciences of robotics
and cybernetics. This (almost)
cybernetic can is able to move
forwards or backwards at the
sound of a handclap. The only
criterium is that the handclap
must be loud enough for the can
to 'hear' it. As the mechanics and
'brain' of the can are completely
hidden from view, your friends
will be completely mystified by
the can's mobility.

How does the can receive this gift of
mobility? The circuit diagram of the
electronic contents of the can is shown
in figure 1. The soundwaves from the
handclap are detected by a crystal
microphone and amplified up to
10,000 times by the chain of gates N6,
N1 . . . N3. The gain of the amplifier
stage can be adjusted by means of PI.
The amplifier section is made up from
CMOS gates - a little known appli-
cation - as one 1C contains a number
of gates and the use of CMOS tech-
nology keeps the current consumption
down to a minimum.

After amplification, the signal is recti-
fied by D1 and shaped by N7. The
resultant pulse performs four operations
simultaneously: flipflop FF1 is reset
and thereby transistor T5 is turned on
via N10. The emitters of T3 and T4
will then be effectively grounded. A
second handclap will have no effect on
FF1 at this stage.

The output of flipflop FF2, on the
other hand, will change state every
time a handclap command is received.
This second flipflop determines the
direction in which the motor, M, will
turn. This change in direction is
achieved with the aid of the transistor
bridge circuit consisting of T1 . . . T4.
When the Q output of FF2 is low,
transistor T3 is turned on and transistor
T2 is turned off (both via N9). Simi-

larly, as the 5 output of this flipflop
will be high at the same moment in
time, T1 will be turned on and T4 will
be turned off (this time via N8). This
means that the motor current will flow
through T1, T3 and T5. When the
outputs of FF2 change state transistors
T2 and T4 (and T5 of course) will con-
duct and the current through the meter
will flow in the opposite direction.

The monostable multivibrator con-
structed around N4 and N5 is also
triggered by the pulse from N7. This
monostable will remove the reset from
the counter, IC1, for about three
seconds. If four pulses (handclaps) reach
the clock input of IC1 during this time,
the Q4 output of the counter will go
high and, via FF1 and N10, transistor
T5 will be turned off. In other words,
after clapping four times in a row, the
can will stop moving altogether.

Finally, the network C9/R12 consti-
tutes a 'power on reset' circuit which
ensures that no current will flow
through the motor when the power is
first applied to the circuit. Similarly,
CIO and R18 reset the counter (IC1)
when the circuit is first switched on.

Mechanical structure
The circuit of the mobile can can be
constructed on two round boards which
have the same diameter as the can. First

r 1980 - 12-39

mobile can

the can must be cut into two parts
(see article on can opening). Figure 2
shows the mechanical construction of
the mobile can.

The 9 V battery can be fixed to the lid
of the can (with double sided tape, for
instance). The microphone can then be
mounted next to it behind the opening
where the ring pull used to be. The two
circuit boards can then be fixed to the
walls of the can slightly lower down. A
small DC motor (6 ... 9 V) is anchored
to the bottom section of the can. A
weighted disc is then attached to the
motor's spindle.

This weighted disc is the only special
component used and, as it is most
unlikely that such a piece of equipment
is laying at the bottom of your junk
box, a word of explanation about its
construction would not be amiss.

The disc can be cut from veroboard,
perspex or other similar material. The
diameter of the disc should be 1 cm less
than the diameter of the can. A hole is
drilled in the centre to accept the motor
spindle (this should be a fairly tight
fit). Four holes are then drilled near
the outside of the disc so that they are
geometrically opposite each other (think
of the points of a magnetic compass).
These holes will then serve as the
anchorage points for the four weights.
The latter can be made from (rolled)
lead and should be secured with nuts

and bolts. The whole assembly should
weigh about 100 .. . 200 g and should
be as finely balanced as possible. It may
require a substantial amount of exper-
imenting and adjusting until the disc
rotates accurately enough. In any case,
it may be necessary to attach small
counterweights to the lid and/or base
of the can to compensate for any
inaccuracies in the final assembly of
the complete unit.

Switch SI is a miniature toggle switch
mounted on the base of the can.
Alternatively, a mercury switch can
be incorporated so that power is applied
to the circuit when the can is laid down.
It is a good idea to give the mobile can
a complete test before fitting the two
sections back together and fastening
them securely in a similar manner to
that described in the can opening
article (candid comments). If every-
thing goes according to plan you can
invite your neighbours around for a
demonstration of the beer rolling home
after you instead of the (more usual?)
other way round. H

din can

L. van Ginderen

The number of possible applications
for noise emitting circuits is absolutely
astounding — far too many to be
listed in the available space. This is
particularly true of the unit described
here. Although not exactly original in
concept, the can is capable of producing
a din which will almost certainly leave
your ears tingling for quite a while.

1

The circuit consists of very few com-
ponents which together form a 'Kojak'
type siren. The majority of the work is
done by two (555) timer IC's. The first
(IC1) generates an audible tone whose
frequency can be adjusted by means of
potentiometer PI. The output of this
timer is fed directly to a loudspeaker.
The loudspeaker however does not emit
a constant tone, as IC1 is modulated by
a low frequency sawtooth waveform
generated by IC2. The frequency of the
sawtooth can be regulated by means of
potentiometer P2. As a result, the
complete circuit generates a frequency
modulated signal which sounds just like
a siren. The pitch can be adjusted with
PI and the modulation rate with P2.

The circuit is compact enough to be
mounted in a can together with the
batteries and loudspeaker and can be
used quite successfully on children's
cycles, skate boards etc. or as an 'anti-
attack' alarm. M

12-40 - i

Our canned genie is extremely lazy. It monostable now returns to its initial
prefers to lie flat on its back in the can, state, the positive going edge at the
or to relax standing on its head, output of N7 turns off T1 and triggers
meditating. It will only allow its master the thyristor. The relay in series with
to place it the right way up and as soon the thyristor will therefore be activated,
as anybody else presumes to try, it will but because C7 is rapidly discharged
make use of its magical powers and through the relay coil the current
knock the can down. Where does it get through Thil will drop very quickly to
these 'magical powers' from? Well, below its minimum holding current. The
actually, the genie isn't that ingenious at thyristor then turns off again and the
all — he's quite simple minded in fact, as relay will drop out.

R. Wenzelburger is the electronics involved. Activating the relay briefly causes a

£$mie in die can

When a beer can refuses to stand
up straight, there must be
something the matter with it. This
obstinate behaviour can only be
due to the presence of an
ingenious genie, rather like the
one in Aladdin's lamp, but smaller
and, unfortunately, minus the
miracles. As is the wont of such
genies, this particular one will
answer to no one but his master
who in this case will not be the
one to call up or release the genie,
but the constructor of this
ingenious can of electronics.

How does it work? When the can is
lying on its side or standing on its head,
the mercury switch SI is open. In this
position (see figure 1) the input of the
'latch' formed by N3 and N4, will go
low. The input to N3 will then be held
low via N4 and R4.

If SI is closed, by placing the can the
right way up, the output of N1 will go
high. This change, however, will not be
perceived by the latch, since D1 and D2
are reverse biased. Both inputs of N5
will therefore be high and its output will
go low. This triggers the monostable
multivibrator formed by N6 and N7
thereby turning on transistor T1. The
pulse duration of the monostable is
determined by the setting of poten-
tiometer PI and will be between 1 ... 3
seconds. During this time capacitor C7
will charge up via T1 and R7. If the

fairly heavy pendulum to move to and
fro, thereby displacing the centre of
gravity of the can to such an extent that
the can topples over . . . the genie
strikes again!

In order to place the can the right way
up without it falling over, sensor S2
must be touched during the process.
The output of N3 will then go low, so
that the multivibrator can not be
triggered. The can will then obey your
command and stand perfectly still.

How to can the genie

After referring to the article on can
opening, use a sharp knife or saw to
remove the base of the can to about
1 cm from the edge. Mind your fingers '. !
There is no need to give the genie a taste

alektor december 1980 - 12-41

for blood!! The ring pull can then be
solidly glued from the inside. To make
sure that the centre of gravity is as high
as possible (so that the can falls over
easily) the battery should be attached to
the lid (with double-sided tape, for
instance). This is illustrated in figure 2.
The pendulum is very simple in con-
struction. A steel rod is glued (or
bolted) to the moving plate of a
(cannibalised) 6 V heavy duty relay. A
piece of lead is then attached to the top
of the rod. The lead should be available
from your local steeplejack, alterna-
tively a fishing weight will suffice. The
required weight of the weight will have
to wait to be established by trial and
error.

Obviously, the centre of gravity is also
affected by the components on the
board (see figure 3). Capacitor C7 in
particular is rather large so it should not
be mounted on the board but should be
placed somewhere near the battery or
the relay.

The sensor S2, must be well concealed.
By far the easiest solution is to cut the
can in two and form the sensor by using
insulating tape to rejoin them (see
figure 2) making sure that the two
sensor halves are sufficiently separated.
A solder tag on both sections is suf-
ficient to provide the connections for
the sensor.

Of course, the genie is its own worst
enemy, since falling over can do it a
great deal of damage. To avoid too
many dents it is a good idea to provide
the can with a few rubber rings (elastic
bands) to soften the worst blows. The

can will then be much less likely to
suffer 'brain damage'.

Remarkably, there is no requirement for
an on/off switch. The quiescent current

of the circuit is extremely low. Since
this depends on the leakage current of
T1, this transistor should be a reliable
BC 558. in other words, no TUN I H

12-42 - elektor december 1980

piping

can ...

electronic wind
instrument

It had to happen, an electronic
equivalent of Scotland's national
instrument, the bagpipes.
Naturally, it is supplied with all
mod cons; an attractive portable,
lightweight case that consists of
none other than a can.

One thing has remained
unchanged — the mouthpiece. The
Scots have developed powerful
lungs to be able to play and this is
still one of the main requirements
for this circuit, modern though it
is.

Those of you who have tried blowing
the bagpipes, know how incredibly
difficult it is to produce any sound that
is vaguely acceptable. Fortunately, the
electronic bagpipes introduced here are
a lot less demanding, but you still have
to blow. The mouthpiece contains an
NTC resistor which is warmed up when
the instrument is blown. It takes some
time to cool off and this creates the
typical bagpipes-effect.

The circuit diagram

Figure 1 gives the diagram of the
electronic bagpipes. The first part of the
circuit forms a VCO and is, in fact,
nothing more than an ordinary oscillator
in which transistor T1 acts as a voltage-
controlled resistor. The control voltage
for the VCO is derived from a voltage
divider network, the ratio of which is
determined by the logic level of the
inverters N1 . . . N4. The high input
impedance of the CMOS inverters makes
it possible to use touch contacts. If the
earth contact is touched with the thumb
and at the same time one of the other
contacts is touched with a finger, the
resistance of the skin will be lower than
the input impedance of the inverter
causing its output to change state. Its
output impedance will then be parallel

to RIO, which causes the voltage at the
junction of the divider R 10/PI +R14to
increase and the frequency of the VCO
to change.

Combinations of several keys are also
possible so that these four switches can
produce a fair number of tones. Differ-
ent values may be chosen for resistors
R9, R11 ... R13 if desired, since these
affect the individual notes. Trial and
error may be the only way to achieve an
individual sound, if that is required. PI
is used to adjust pitch.

The NTC resistor in the mouthpiece
together with R17 also constitutes a
voltage divider. The voltage across the
NTC is compared by IC2 to a fixed
value (persettable with P2). When the
instrument is not blown, the output
voltage of IC2 drops taking the collector
voltage of T2 to zero. This effectively
blocks the oscillator output and so the
loudspeaker will not produce a tone.
When the instrument is blown, however,
the NTC resistor heats up and the
voltage at the junction of R17 and R18
drops. When it becomes lower than that
at the non-inverting input of the opamp,
the output voltage will rise to 9 V. As a
result, the oscillator output will now
control the loudspeaker. When the
instrument cools off, the voltage across
the NTC will drop again, causing the

>r december 1980 - 12-43

sound to subside after some time. The
circuit's sensitivity can be affected by 2
changing R22, if necessary.

The case

Figure 2 shows how the circuit is
mounted in the can, The lid has been
sawn off and holes have been drilled in
the lower side for the loudspeaker. On
one side of the can there are the touch
contacts and on the other the thumb
key. The sensors may be drawing pins.
The printed circuit board and the
component overlay are illustrated in
figure 3. This fits the can exactly. The
on/off switch is mounted in the lid
together with a socket for a standard
jack plug. The NTC resistor is connected
to the pins of the jack plug. A length of
pipe is glued onto it and provided with
an opening just in front of the NTC.
Figure 2 is a drawing of the mouthpiece
in greater detail.

Now, how to play the instrument!
Switch the supply on, place your thumb
on the earth sensor and one or more
fingers on the others. Then blow as hard
as you can until a sound is produced.
After a while it's good to remember that
the blow threshold can be preset with
P2. Please, don't take it to Glasgow! M

ill look like, once the instrument is built and

amber 1980

simple

control

Remote control systems have
gained a foothold in all walks of
life, whether it be for reasons of
comfort or practicality. Probably
the best known and most used
system is the remote control of
domestic television sets, but of
course there are many other
possibilities. Elektor has
concerned itself with the subject
on several occasions in the past
and is pleased to present yet
another variation on the theme —
one which will benefit the ill and
disabled in particular.

F. Kasparec

Conventional remote control trans- The transmitter
mission units are often adorned with a

large array of buttons or switches. Figure 1 gives an impression of what
meaning that a certain percentage of the transmitter unit looks like. It
operational errors cannot be avoided, is constructed from a standard sized
The alternative introduced here requires beer or soft drinks can with the text
only one pushbutton, thereby enhanc- 'Source I . . . Source IV' written on the
ing its security and ease of operation outside to indicate the four possible
considerably. All that is required is a functions. Naturally, the wording could
glance at the descriptions on the can be altered to: lights, television, radio,

remote

corresponding to the four possible heater, for instance, if required. If the

functions, and then press the button. transmitter is only to be used to control

The message is transmitted ultra- four light sources, the familiar dimmer

sonically for two reasons: 1C, S 566B, could also be incorporated

1. the G.P.O. and the Home Office are into the lighting system. The single

very particular about the use of radio pushbutton can be labelled 'PRESS' if

waves. desired. So much for the exterior.

2. if an infra-red system was used the The inside of the remote control
energy consumption would be very transmitter is very compact. In addition
high unless a lens was incorporated. to the transmission electronics there is

Figure 1 . The transmitter for the remote control system can be incorporated (at a pinch! I in a
'standard' can. All the components should be firmly mounted, with the exception of course, of
the revolving encoder disc.

elektor december 1980 - 12-45

2

four channels is to be operated.

an optical encoder which selects one of
the four functions to be transmitted.

The entire circuit is shown in figure 2.
The encoder consists of two photo
detectors (D1/T1 and D2/T2) as well as
the encoder disc (see figure 3). Two
LEDs are used as light sources and
virtually any type of photo transistor
can be used for the receptors. An
interesting prospect would be to incor-
porate 'ready made' photo detectors,
where the LED and photo transistor are
already integrated into a compact
assembly.

The encoder disc is positioned in the
can in such a way that it can rotate
freely between the two photo detectors.
The position of the encoder disc will
then determine whether both photo
transistors are turned on, turned off, or
whether just one of them is turned on.
The two-bit information thus obtained
can then be deciphered by the following
demultiplexer. This is accomplished
with the aid of inverters N1 . . . N4 and
the AND gates N5 . . . N8.

The decoded 2-bit information is then
passed to the MOS switches SI . . . S4
where SI corresponds to position A of
the encoder disc, S2 to position B and

The frequency of the astable multi-
vibrator constructed around IC4 is
determined by the capacitor C3 and by
the resistance value connected between
pins 2 and 3. This resistance value will
depend on which of the four switches,
SI . . . S4, is closed. If, for instance,
switch SI is closed, the combined value
of PI and R4 will determine the fre-
quency of oscillation. As the preset
values of potentiometers PI . . . P4 will

each be different, there will be four
different modulation frequencies avail-
able at the output of the AMV (pin 11).
The output signal is fed to the modu-
lation input of a 555 timer (IC5) via a
low pass filter. The timer is also
connected as an astable multivibrator.
The frequency of the second oscillator
is modulated by the signal from the
first. The resultant (FM) output signal
at pin 3 of IC5 is fed to the ultrasonic
transducer via transistor T3 and the coil
L2.

The on/off switch for the unit is a
pushbutton (S5). It has two functions.
It connects the power supply to the
circuit and simultaneously causes the
desired function to be transmitted. This
guarantees that current is drawn from
the power supply only when the remote
control unit is actually operated.

With regard to the receiver, the circuit
published in the 'remote control slide
projector' article (Elektor October 1980,
page 1 0-24, figure 3) could be a possi-
bility. Depending on how it is to be
used, it may have to be modified here
and there.

A few other hints

The encoder disc can be made from a
piece of relatively thin perspex with a
diameter of not more than 5 cm. The
necessary markings can be made up
from black self-adhesive tape and stuck
on to the disc (see figure 3). The
encoder disc must be mounted between
the two photo detectors in such a way
that it can turn freely and produce the
correct 2-bit information.

Although the disc is mobile, it should

Figure 3. For two-bit information the
construction of the encoder disc will be
relatively simple.

not be able to turn when the can is
swivelled horizontally on its axis. To
ensure that this does not happen, the
hole drilled in the disc near to the edge
should be used as a fixture for a counter-
weight. The weight will then keep the
disc in a fixed position even if the can is
swivelled rather abruptly. Suitable
counterweights can be made up from
washers held by a nut and bolt. When
the can is rotated the counterweight
should not touch the board at any
position!

The printed circuit board and the
battery should not be allowed to roam
around loose inside the can. It is best to
pad the inside with foam rubber or
some other suitable material. The
constructional drawing of figure 1
should answer any other queries regard-
ing the contents of the remote control
transmitter.

Finally, the markings on the outside of
the can. It is obviously important that
each one should correspond to the
command executed. Once the trans-
mitter is complete and the receiver has
been installed, you're ready to go: just
look at the function you require and
press the button. H

This is an outdoor game for two players,
both armed with a bottle of water and
a ball. At a certain distance from each
other they start throwing a ball at each
other's bottles. Every time the bottle
falls over some water will run out and so
the first to be left with an empty bottle
is the loser. A simple, if rather wet
game. The author, however, has managed
to develop a dry version using elec-
tronics. The bottle is now a can contain-
ing a few squeaking bits and pieces.
When the can is knocked down, instead
of water coming out of it, it will squeak.

water can

When it has been knocked over so many
times that if it had been a bottle it
would now be empty, it emits an inter-
mittent squeal to indicate the end of the
game.

The circuit

A 214 inch loudspeaker and a 9 V bat-
tery are mounted in the can, as well as
a small printed circuit board incorpor-
ating a very simple circuit, more of
which will come to light later.

The can also contains a switch that
closes when the can falls over. This con- I
sists of a length of wire with a metal
weight suspended on it, which consti-
tutes one contact. The other contact of
the 'moving switch' is formed by the
inner wall of the can.

The diagram shows the weight can com-
bination in the top lefthand corner. As
soon as the can is knocked down and
the weight hits the inner wall of the can,
the oscillator built up with N4, R3 and
C3 is started. The squeak produced is
amplified by T1 and made audible
through the loudspeaker.

When the can is on its side, the electro-
lytic capacitor Cl is charged via PI and
03. After the can has fallen over a
couple of times, the voltage across Cl
will be so high that the oscillator around
N3 (frequency about 3 Hz) will be
started by N1/N2.

As soon as the can is placed the right
way up, oscillator N4 (which was
squeaking all the time the can was on
its side) will be modulated by the 3 Hz,
so that the intermittent loser tone will
sound in the loudspeaker to indicate
that 'the bottle is empty'.

On average the game lasts according to
the setting of PI. SI acts as a start
switch which enables Cl to be dis-
charged quickly and so bring the circuit !
to the beginning of the cycle. By the
way, it is best to mount this switch on
the top of the can, as otherwise it might
get depressed when the can falls over.

If you think the can is rather a small
target, another can be mounted on top

elektor december 1980 - 12-47

canine

Pets can be found in any shape or
size as the recent pet rock craze
in the USA testifies. Some people
find warmth and companionship
in a can (it takes all sorts . . .).
This article is for those readers
who wish to protect their pet. No
longer shall your can be
completely defenceless against
the rough and tumble of everyday
can life and its most feared
enemy, the young human male.

T. Stokes

The canine defence is in two parts,
first line defence and second line
defence. The first is in fact operating
permanently in the form Of a flashing
light. This will enable you to locate
your pet can at night and has the added
bonus of indicating to other humans
that your can is alive and well. This part
of the circuit uses the well known
LM 3909 to flash an LED mounted in
the top of the can. The value of timing
capacitor Cl can be varied to alter the
flash rate if required.

The second part of the circuit is an
alarm. When the can is tilted or in any

defence

other way disturbed the mercury
switch MS will activate the monostable
formed by N1 and N2. This will in turn
start the oscillator (N3/N4) and a
continuous tone will be produced for
about ten seconds before switching
off. This timing is determined by the
values of R3 and C2 while the fre-
quency of the oscillator depends on R4

Figure 1. The first pan of the circuit uses the
LM3909 LED flasher 1C.

ponents can of course be altered to suit
the individual reader, although the din
produced by the values given should
be enough to protect most cans.

A supply voltage of 4.5 volts is re-
quired and this can be supplied by
means of three HP7 batteries connected
in series. Since the quiescent current
drain is very low an on-off switch is
not necessary and the batteries can be
carefully soldered into the circuit with
short connecting leads.

Construction

To fit the circuit the bottom of the can
must be opened. First, try to persuade
your pet that it is in his/her own interest
that this operation be carried out. The
article on can 'surgery' should be
followed carefully with particular atten-
tion paid to the anaesthetic. This can
take the form of medicinal liquids,
brandy for instance, with the added ad-
vantage that it does not necessarily
have to be wasted. Two or three 'rinses'
could suffice although some readers
may prefer more.

The circuit may be assembled on
Veroboard (or similar) and fixed to the
inside of your pet with 'Blu-tack'. None
of the components are in any way
critical and are available from most
suppliers.

An 80 ohm loudspeaker will give the
maximum output but any speaker will
do providing it will fit inside the can.
The total resistance of the speaker and
R6 should be about 100 ohms. Remem-
ber to drill a few holes in the top of
the can before fitting the speaker.

For the owners 'who have everything',
two (or even more) complete circuits
could be fitted into your pet with one
speaker at the top and the other at the
bottom, if you'll pardon the expression.
On a serious note, the completed can
without the flashing light, would make
a novel 'burglar alarm' if placed inside
the bedroom door. But of course, it
could have hundreds of uses ... H

2

Figure 2. Any type of mercury switch can be used in the alarm circuit. Care should be taken to
protect it from damage inside the can.

12-48 -ele

iber 1980

electronic garden gnome

dedronie gwdoi gnnne

For the keen gardener, a moisture
indicator with a LED display
could be a very useful aid. It is
simple and cheap to build and
can be used anywhere including
the greenhouse. For the gnomeless
garden it can even be equipped
with a rod and line . . .

W. Holdinghausen

It would appear that garden gnomes are
a disappearing commodity in the
contemporary garden. A suggestion that
the majority left in the early seventies
with gnomention of where to was
dismissed. However, the replacement
presented here is far more progressive
than its old counterpart and of course,
it will not budge an inch (unless
kicked).

Using cans for aesthetic purposes is by
no means new. The idea was one among
many that brought Andy Warhol to
fame (and riches) in the sixties. For the
purposes of this article, the can is ideal.
It is water-proof, readily obtainable
and can be purchased full of all kinds
of beverages. All it requires is emptying
to make the perfect garden gnome.

Our gnome contains an electronic
I circuit (of course) designed to measure
the moisture content of the top layer of
soil. Indication to the gardener is by
means of LEDs allowing him to see at
a glance if watering is required. If
painted in frightening colours, the can
will also serve to scare off birds provided
enough of them are used. Even better
would be to include an acoustic circuit,
examples of which may be found else-
where in this issue. Although, whoever
heard of a frighteningly coloured gnome
with lights on its face measuring moist-
ure while scaring birds and fishing?

The circuit diagram

Enough about gnomes and on to the
circuit which uses an analogue switch
as the heart of the moisture detector.
Between the two sensor probes, the soil
will represent a resistance Re with a
value dependant on its moisture con-
tent. The potentiometer PI together
with Re constitutes a voltage divider
with its centre at the control input of
the CMOS analogue switch ESI.

As the soil dries out, the value of Re
will rise causing an increase in the con-
trol voltage of ESI. When this reaches
a level of about two thirds of the supply
voltage, ESI will close and be latched

Figure 1. The electronic garden gnome is a soil moisture sensor using LEDs that flash when the
garden needs watering.

2

Figure 2. The drawing shows an up-to-date garden gnome prototype. The facial features are
provided by three light emitting diodes and a switch.

by D1. At the same time ES2 and ES3
will also close lighting LEDs D2 . . . D4.
The capacitor Cl will then charge via
ES2 and R2 to about 6 volts. When this
level is reached, ES4 will switch the con-
trol inputs of ES2 and ES3 to zero until
Cl is discharged again via R2, R3 and
D2. All this causes the LEDs to flash
on and off to a three second cycle.

If the garden is now watered, the
increase in moisture content of the soil
will lower the resistance of Re thereby
switching off ESI.

The threshold at which dryness is
indicated is adjustable by PI. The
correct level depends on the type of
soil, the distance between the probes
and the amount of growth in the soil.
Adjustment can best be determined by
experiment. Switch SI is the test button
and pressing this will show if the water
can is needed.

It's amazing that it takes science (and
Elektor) to find out that an ohm or
two in the garden can protect your
plants from withering. M

Low cost 2716 EPROM
programmer

A low cost programmer for use with INTEL
2617 EPROM memories has been introduced
by Technova Developments Ltd. The unit is
particularly useful for the development
engineer wishing to modify an existing
programs.

reference, a high stability thin-film R-2R
ladder network, and a 0 lo 2mA current
output for 0 to 10 V outputs. Fast setting
time of 200 ns typical and low power con-
sumption of 200 mW typical are combined in
a 16 pin DIP package. Non-linearity grades are
0.1% maximum. 0.2% maximum, and 0.3%
F.S. The DAC-101 operates over a supply
range of s6 V to ± 18 V and can directly
accept DTL and TTL logic inputs; CMOS
logic are easily adapted.

The high reliability of the DAC-101 is en-
hanced by 100% burn-in, 100% stabilization
bake. 100% centrifuge. 100% temperature
cycling, and 100% fine and gross leak per
MIL-STD-883B.

Precision Monolithics Incorporated,

1500 Space Park Drive.

Santa Park Drive,

Santa Clara,

California 95050.

Tel.: 14081 246-9222.

TWX 910338-0528.

Cable Mono

Park Street Industrial Est.,
Aylesbury,

Buckinghamshire, HP20 1ET.
Tel.: Aylesbury 102961 20441.
Telex: 83570.

The length of the programr
crystal controlled for reliable pr
Technova Developments L td,
Blofield Heath ,

Norwich, Norfolk. Nr 13 4SF..
Tel.: 106031 712288

10-bit, low cost, high speed,
digital-to-analog converter

reliability, and high speed at low cost
applications having a limited temperat

e eprom. Jubilee cases

West Hyde Developments, the Aylesbury
based specialists in electronic packaging, have
recently introduced the 'Jubilee' range of
cases. There are two new cases in the range
highly suitable for visual display units, video
terminals. TV games, keyboards and micro-

41 61 2 style socket connector

The latest addition to the range of 3M Scotch-
flex insulation displacement connectors is the
3553 64 socket connector which has been

DIN 9 41612 specification. It has two rows of
contacts with • 100" by • 100" spacing.

The 3553 is available in versions with or

tabs are also available if required.

A new locator plate - the 3443 -64S has
been designed specifically for both versions of
the 3553 connectors. It is supplied with a
shim already fitted, so that the Scotchflex
Assembly Press operator can change the
locator plate without having to adjust the
height of the press. The plate fits all
Scotchflex assembly presses.

moulding process, both cases are easily
machined using ordinary tools. The standard
finish is textured light grey, with a textured
black keyboard plate.

VDU TVT1 Micro-Computer Enclosure
A case with a flat top behind the keyboard
panel. Top height 35 mm, overall height
137 mm. The base moulding has ribs and

slots, and a 112 mm diameter fan opening
(four fixing holes on a 148.5 mm diameter

12-50 -elektor de

market

The 3553 socket connector is fully compat-

ible with standard DIN 41612 style header-
connectors available throughout the industry
and complements the other Scotchflex
DIN 41612 style, the 3332 with • 100" by
• 200" spacing which is already available.

3M United Kingdom Ltd..

3M House,

PO Box 1,

Bracknell.

Berkshire RG12 1JU.

Telephone: Bracknell 103441 58436.

(1713 M)

SRQ series BCD and BCHEX
miniature programme switches

Ambit's data products division now offers a
neat solution to 'on-card' programming via
BCD and BCHEX with a range of switches for
either vertical or horizontal mounting. For

far more satisfactory than OIL switching - as
well as being cheaper.

the standard 13 amp mains socket. It contains

a vacuum-impregnated double-wound trans-

former, giving 40 watts output at 24 volts.
Low-voltage soldering irons, heated wire
strippers, etc. up to 40 watts loading may be
wired into the special detachable output cleat
of the DOHMplug. which is then plugged
directly into a 240 volt mains socket. There
can be no danger as the DOHMplug provides
only low-voltage output. The transformer
windings are protected by a 250 mA cartridge
fuse. A particularly useful application is the
operation of the LITESOLD LE40 24 volt
electronically temperature controlled soldering
iron, where the use of the DOHMplug avoids
the need for a much larger and more expens-
ive power unit, and provides in effect a totally
safe and portable 'mains' electronically
controlled iron, as illustrated in the photo-
graph.

Light Soldering Developments Limited.
Spencer Place.

97/99 Gloucester Road.

Croydon. CRO 2DN. England.

Telephone: 01-6890574.

Telegrams: Litesold Croydon.

Telex: 8811945

(1718 M)

New pocket-size, hand held-box

A pocket size, hand-held box, ideal for
housing remote control handsets, instruments

Brentwood.

Essex. CM14 4SG,

Tel.: 10277) 230909.

Telex: 995194 Ambit G.

(1 719 Ml

Mains plug/transformer unit

Light Soldering Developments Ltd. are
introducing a very useful combined mains
plug and transformer unit, allowing the use of
low-voltage soldering irons directly from the
240 volt mains.

The DOHMplug has a tough moulded
polypropylene case, with integral pins to fit

etc., is the latest addition to the Vero range of

plastic enclosures.

It incorporates a 20 x 50 mm cut-out slot
which may be used for fitting a VDU panel or
switches and a 1 2 x 35 mm recessed panel for
labelling. A circuit board 71 x 107 mm can be
accommodated in the top section of the two
part body by the use of four self-tapping
screws, whilst the bottom section will house a
board of 56 x 105 mm. The two sections snap
together and are secured by four screws which

enter through the base, when this is done

rugged strength and at the same time soundly
seals the inside circuitry. There is an integral
battery compartment which will accept a PP3

or a nickel cadmium stack of 25 x 45 mm, it

has a slide-off cover which allows easy access
for insertion and removal of batteries.

This attractively styled enclosure is moulded
in high-impact polystyrene which has a
textured, non-slip finish to aid handling and

Vero Electronics Limited.

Industrial Estate.

Chandler's Ford,

Eastleigh,

Hampshire.

S053ZR

Telephone: 10703) 66300

(1 716 M)

High profile Dl Lswitch

A new 3 pole way, high profile DILswitch has
been added to the ERG DIL 16 range of dual
in-line switches. The body measures 20.5 long

protrudes above the main body by 2.6 mm.

Aimed at front panel switching applications
where space is at a premium, this new
component has gold plated contacts with
ratings of 240 V a.c. 2 A (non switching),
30 V 0.25 A (switching). Typical initial

an insulation resistance of 100 GO, while
operational temperature range extends from
—55 to +100°C. Operating force is 250 grams
(typical). Manufactured to meet BS 201 1

specifications, life expectancy within typical
rated load is < 20,000 break/makes.

Erg Industrial Corporation Limited,

Luton Road.

Dunstable,

Bedfordshire. LU54LJ,

England.

Telephone: 0582-62241 (7 lines).

Telex: 82349

(1727 M)

J

materials with established techniques. A
major factor in achieving the high performance
s the use of samarium-cobalt magnetic
material. When coupled with Portescap's
ironless rotor design, gold-alloy brushes and
sintered-bronze bearings, one achieves not
only high efficiency but also high reliability
and maintenance-free operation.

Rortescap ( U.K .) Limited,

204 Eiger Road,

Reading, RG2 ODD.,

Tel.: Reading (0734) 861485/6/7/8,

Telex: 847661

Rotary speed convertor

Gould Instruments Division. Designated
Gould Model 13-4618-30, it is ideal for
testing all types of internal combustion
engines, electric motors, variable-speed drivers,
and other rotating machinery.

For convenience and reduced chance of error
in automotive applications, a multiposition
switch allows the operator to select the
number of cylinders or number of fan blades
on an engine being tested. Selecting 2, 4. 6 or
8 cylinders on this switch causes the output
to read directly in revolutions per minute
when ignition-coil inputs are used. Selecting
1,2,3, 4. 5, 6 or 7 fan blades gives the
correct rev/min output when a magnetic or
optical pickup senses that number of impulses
per revolution from a rotating cooling fan or

Gould Instruments Division,

Roebuck Road.

‘Simple to program' 1C test
system from MIT(E)

The Instruments Division of specialist
suppliers MIT (Engineering) Ltd has extended
its range of 1C Test systems by introducing a

Peripheral interfaces allow the ICT-4100 to I

Teletypes, Paper Tape Readers/Punches ar
VDUS, etc., so storage of programs can be
any media the user desires.

(Engineering) Limited.

Greenhiii Industrial Estate,

Ridding,

Derby DE55 4DA,

Telephone: Leabrooks 604411 ,

Telex: 377008

New sony radio turns science
fiction into science fact

Sony technologists have turned science fict
into 'science fact' for the average ra
listener with the development of
ICF-2001 , which the company describes
'the most advanced portable communicati
receiver in the world'.

programmes from Australia to
foreign language services from ■
corners of the globe, ship-to-shore,
and citizen band transmissions, as

COMPUTERS AUDIO RADIO MUSIC LOGIC TEST GEAR CB GAMES KITS

26th Nov - WEDNESDAY - 10am-6pm
27th Nov - THURSDAY - 10am-8pm
28th Nov - FRIDAY - 10am-6pm
29th Nov - SATURDAY - 10am-6pm
30th Nov - SUNDAY - 10am-4pm

COMPONENTS DEMONSTRATIONS SPECIAL OFFERS MAGAZINES BOOKS

Royal Horticultural Halls Elverton Street _
Westminster London SW1
November 26-30 1980
It’s all at Breadboard ’80

This is the exhibition for the electronics enthusiast. From
November 26 -30 there is only one place in the universe for the
electronics enthusiast to be - Breadboard '80, at the Royal
Horticultural Hall in London. The majority of leading companies will
be exhibiting, including all the top monthly magazines in the field.

There will be demonstrations on most stands and many feature
special offers that are EXCLUSIVE to Breadboard!

All aspects of this fascinating field are catered for, from CB to
home computing, so whether you want to buy a soldering iron or a
synthesiser — or just keep up to date with your hobby — don't
miss Breadboard '80.

AVOID QUEUES - GET ADVANCE TICKETS

Send to: Advance Tickets. Modmags Ltd.

145 Charing Cross Rd. London WC2H 0EE

jdent/Child/OAPI

junior

computer

For anyone who wishes to become
familiar with (micro)computers,
this book will mean the start
of a fascinating hobby. Both
beginners and experienced
hobbyists now have the oppor-
tunity to build and program their
own personal computer at a very
reasonable cost.

m

mill

m

Hi

il